Graphene and NPL: new standards for a maturing industry

Collaborations Graphene Engineering Innovation Centre Research 19th March 2021

Dr Andrew Pollard is the Science Area Leader of the Surface Technology Group at the National Physical Laboratory (NPL), which is an affiliate partner of the GEIC and a long-time collaborator with The University of Manchester on standards around 2D materials. Andrew tells us about his research at NPL and why its latest standard is so important for the graphene supply chain.

Andrew, please outline your role at NPL…

After finishing my PhD on graphene and boron nitride, I moved to NPL where I have been developing the metrology and standardisation of graphene and 2D related materials, which are required to enable industry working with these materials.

Graphene is a term that’s widely used, but there are lots of different forms that are on offer and other types of 2D materials. Our work is about understanding the measurement of these nanomaterials – their structural and chemical characterisation – reliably and reproducibly. Then, working through organisations like ISO, we ultimately set an international standard so that everyone can measure these properties in the same way.

The reason that’s important is that it provides industry with confidence in the supply chain and enables the global 2D materials industry to grow.

For those who don’t know, how would you define metrology?

Metrology is the science of measurement – not just taking measurements, but understanding the uncertainty around that measurement, as every measured value will have an uncertainty associated with it that provides an idea of the confidence in that value. We can then try to reduce that uncertainty to produce the most precise and most accurate measurements you can, or at least verify what the uncertainty is for that measurement.

In some cases you might be developing a quality control process, where a large amount of uncertainty is fine as long as the measurement is quick to perform, but you need to verify what that uncertainty is and describe its traceability by relating it to a known reference.

Do you think there was enough focus on standards in the early days of graphene, when some early adopters were put off by variable quality?

I remember, maybe 10 years ago now, talking to companies about how standards were important, but they do take time to put together. So I don’t think graphene is unique in this – although maybe because it’s moving at such a fast pace it makes it difficult for standardisation, which runs at a slower pace because of having to reach international consensus. A good standard is always based on metrologically sound research and that takes time as well.

It’s safe to say that industry would definitely have liked standards sooner on graphene – but it’s taken the time that you might expect to produce something sufficiently robust and internationally agreed.

So where do you think we are with graphene along the line of evolution from emergence to mature industrial commodity?

For standards, we look across that evolution starting at the very beginning with terminology – so everyone’s speaking the same language. For graphene, that standard came out in 2017. And then it’s really about measurement standards. People think of standards in different ways, but we’re not talking yet about defining product quality (ie. how good your product should be).

We’re still talking about a document that says ‘this is how you should measure your material’ so that everyone does it in the same way and can then compare technical data sheets.

The new standard is the first ISO measurement standard that details how you to measure your material – as a powder or a liquid dispersion, as opposed to single-layer sheets on a substrate – and say precisely what you have: namely if you have few-layer graphene and how much.

I think we will see future standards coming out that would be application-specific, but we’re still fairly early on in the standardisation process. If you look at carbon nanotubes, they’ve been around a lot longer than graphene but there are still new standards being developed in that area.

Will a single method of measurement disadvantage people who are maybe using kit and techniques that don’t align with that methodology?

The important distinction is in what we would term the ‘metrology pyramid’, where at the top you have a few entities who can measure a property very accurately and precisely and at the bottom a very accessible measurement with much greater uncertainty.

So if you look at industry – a lot of the time they’ll have quality control processes that they themselves have developed. That’s what you’d expect them to be doing on every batch of material to make sure that the material is good enough, but at the same time with typically a high level of uncertainty.

This standard is more at the top of the pyramid, so saying: ‘how can we measure as accurately and precisely as possible?’. You wouldn’t expect every producer to be able to measure every batch in this way as it’s likely to be an expensive process.

If I give the example of a resistor – generally, most people who want to measure resistance in a lab will grab their multimeter and look at what the resistance is and that’s fine. Maybe there’s an uncertainy of 10-20%. But there are places in the world that can measure resistance to an accuracy of nano-ohms.

Now, you don’t need to do that when you’re just in a lab with an ordinary resistor, but ultimately to make sure that you have a chain of traceability, to know that multimeter has an uncertainty of X%, you need to have increasing levels of accuracy and precision to verify against, as you travel up the metrology pyramid, until you get to the most accurate at the top.

So this standard is about setting the top of the pyramid and it has been designed to use the techniques that are generally used in the field of 2D materials. There’s no niche technique here that only one place in the world can use. It’s just to make sure that ultimately everyone is comparing apples with apples.

Does this new standard apply to measurement of all grades of graphene – from single-layer CVD down to bulk powders?

What we’re trying to do within the ‘Nanotechnologies’ Technical committee in ISO (TC229) is to try to have a family of standards so you can get a complete understanding of the materials being used in industry, ensuring different measurement standards will apply to different forms. So we have split it essentially into flakes, which are generally in a powder or liquid dispersion, and into sheets on a substrate, ie. CVD-grown [chemical vapour deposition] graphene.

These standards are separate, so you can choose which of those two forms you are dealing with, because the measurement techniques might be the same but the ways in which you apply them are very different, therefore you need to have different standards and different protocols.

In the NPL Good Practice Guide 145 that the NPL published jointly with The University of Manchester in 2017, we make this distinction of the two different forms, with separate sections for each.

There are then also different measurement standards being developed for different properties, namely structural properties like number of layers, lateral size, defects, as well as chemistry: what’s the amount of oxygen in here? What’s the level of contamination or metal impurities?

What specifically does the new standard focus on?

The new ISO standard looks at the structural properties of flakes in powders and dispersions. Now, that’s not an accident. We’re generally seeing the bigger part of the market as it stands involving these forms. That seems to be where most of the demand is from industry. The first questions are: what’s the structure? Is it few-layer graphene? What’s my yield? That’s the reason this standard has been prioritised and is the first to be published.

Typically, developing a standard from initiation through to publication can take somewhere between two and five years because you’re reaching international consensus. In this case, more than 30 countries are part of the technical committee on nanotechnologies, which is where graphene sits. They agree a standard should be developed and it goes through different drafts that are balloted and improved at each stage. Again, this is a reason why producing a rigorous international standard, that everyone can agree on, does take several years.

So in this case, what has been agreed?

In this case we have defined the measurement techniques you would use because it’s not possible to record the measurands we are interested in with just one technique – we’re detailing the protocols for Raman spectroscopy, transmission electron microscopy, scanning electron microscopy, atomic force microscopy and gas physisorption using the BET method.

Ultimately, the idea of the standard is to help industry. Rather than buying hundreds of samples from companies around the world, they can instead look at the datasheets of commercially-available materials and compare them because they’ve been measured the same way using the same standard.

If they know they need a certain kind of material with certain properties they can then limit the samples they need to investigate down to a much smaller amount of materials. That’s our aim.

In what ways do you think the graphene supply chain has improved over the years?

I think we’ve seen a real maturation of the industry over the last 10 years because we’re now seeing large scale production. In the past, larger companies who would be the end-users were saying ‘until you can prove that you can make tonnes of this stuff then it’s not worth us even looking at it’. We’ve got that now, but one of the other problems companies were finding was batch-to-batch reproducibility and that becomes a bigger issue as you scale up.

We talked to lots of companies in the past who said ‘we got some material one time from a company and it worked well and so we thought we’d look at it in greater detail but then further down the line we couldn’t repeat the same results’.

The industry has matured since then, but we do still need reliable quality control methods. And if you can develop quality control processes that are simpler, more rapid and reproducible, then they can be used by someone at a technician level, rather than finding your local professor in that technique. This is really important to keep costs down for companies.

What’s the quality control procedure for your own standards? Is there a two-way information street between you and those using your standards?

Essentially, yes. There is a bit of a feedback loop. If you’re trying to understand uncertainty for an international standard, you’re going to have places all around the world measuring and using that standard, so things such as instrumentation and different users cause a greater uncertainty than if it’s one person in a lab doing the same thing over and over again.

Yes, you can work out the standard deviation for that one example and that’s great but to really understand the uncertainty that you’re looking at when comparing different technical datasheets, you need to compare how the values recorded will vary across many different labs with different instrumentation.

So what we do is international inter-laboratory studies, where different users with different instruments essentially measure the same material with the same methodology. Hopefully this year we will see at least one publication published on this work that is being done within VAMAS.

What’s VAMAS?

VAMAS is an international entity that orchestrates the international inter-laboratory comparisons for advanced materials. There are lots of different technical working areas and one of the newer ones is graphene and related 2D materials (TWA 41).

So we’re doing these studies within VAMAS TWA41 and they are on the techniques being developed within these ISO measurement standards, through these you can identify any measurement issues. Are there any outliers? Why are they there? Is there a systematic issue with the measurements? We then need to make a comment on that in the standard to try to prevent that happening.

But it doesn’t stop there. When a standard is published, they then come up for review periodically – with the time period depending on the level of the standard – and the countries involved get to vote on whether they think they need to review the standard. That’s common practice and you would expect revisions to happen in a newer field such as graphene.

It might be that you have a very good standard but something is discovered in the future and you improve it and reduce the uncertainty yet further. The science keeps moving on and improving.

Find out more about NPL’s work in graphene and 2D materials.

If you are interested in partnering your business with Graphene@Manchester to help with the rapid prototyping and scale-up of graphene applications and products, fill in the contact form and a member of our business engagement team will be in touch.