FOR years, and were like most other physicists, labouring away relatively unknown to the wider world. The pair studied crystals in their lab at the near Tokyo, Japan.
Then, almost overnight, they hit the big time. They had been growing a cubic crystal form of boron nitride that has the same three-dimensional structure as diamond. One day, out of curiosity, they investigated another type of boron nitride crystal that sometimes grew as a by-product in their lab 鈥� a flat, two-dimensional form.
Advertisement
With it, they inadvertently struck gold. That鈥檚 because, around this time, another 2D substance was starting to make waves. Graphene, formed of a sheet of carbon just a single atom thick, was dubbed a 鈥渨onder material鈥� due to it being a great conductor, stronger than diamond and lighter than paper. An influx of graphene research began, trying to make the most of this stuff.
The problem was, to study graphene, you need something very flat with just the right properties on which to mount the wafer thin sheets. The solution, it turned out, was the very by-product crystals Taniguchi and Watanabe had been investigating.
Their high-purity 2D boron nitride crystals are, by wide consensus, the world鈥檚 best. Today, what was once a waste material is supplied to everyone in the graphene field to enable groundbreaking research and the two scientists are co-authors of more than 1000 studies. They told 麻豆传媒 how they honed their craft, found themselves at the centre of a materials revolution and became the world鈥檚 most in-demand crystal growers.
Anna Demming: Let鈥檚 start with the basics. What is a crystal?
Kenji Watanabe: A crystal is a solid where the atoms are arranged in a repeating pattern. This affects how the electrons behave, which can lead to interesting electronic and optical effects.
What were the first crystals you studied?
Takashi Taniguchi: I started out working on making diamond. Diamond is not only a gem, but it is also a very hard material and a semiconductor 鈥� something that only conducts electricity if the electrons [in it] have enough energy to 鈥渟witch on鈥�. Diamond has a wide bandgap 鈥� the minimum amount of energy needed for electrons to move through the material and create a current. This kind of behaviour is useful for devices, so companies all over the world are synthesising diamond, not just for jewellery, but for semiconductors.
How did you go from diamond to what you are now known for, boron nitride?
TT: Diamond is made of carbon, but next to carbon on the periodic table is boron on one side and nitrogen on the other. This means that boron and nitrogen, when put together, combine to make a similar structure to carbon, called boron nitride. I realised that while there were many research papers reporting on diamond, there was not so much on boron nitride, so there was a lot of space for further study.
KW: I had been studying the properties of semiconductors for high-power electronics. Boron nitride and diamond looked promising, but it was difficult to find large crystals of these materials.
What makes a good crystal?
TT: Diamond should be colourless. But man-made diamond [using high-pressure techniques] is always yellow. This is because there are nitrogen atoms in it. It鈥檚 the same situation with boron nitride: if we want high-quality boron nitride, we need to remove the impurities, such as carbon and oxygen. So high quality means high purity.
How did you hone your crystal-growing skills?
TT: We learned a lot about growing cubic boron nitride from diamond synthesis. Graphite, loosely bound layers of carbon arranged in a hexagonal crystal pattern, converts to diamond with the right chemicals and high pressure. The same thing happens for boron nitride.
We started with hexagonal boron nitride, which is commercially available as a powder, and using a hydraulic press, a very big one, applied around 30,000 tonnes of pressure. We needed to find the right solvents to dissolve the hexagonal boron nitride and then create cubic boron nitride crystals. Once we did, it took a long time to grow the crystal 鈥� about five days. I tried many materials by trial and error and finally found putting barium in the solvent was very useful. This was the key.
Today, the crystals you work on most of the time aren鈥檛 cubic, but arranged in a flat, hexagonal pattern. What made you switch your attention?
TT: Originally, this was sort of a by-product. When we grew the cubic boron nitride, hexagonal boron nitride was also produced, only much higher purity and with a larger crystal size than the commercially available starting powder.
KW: I became curious because, at the time, there weren鈥檛 many articles reporting the bandgap and the luminescent nature of hexagonal boron nitride.
How did that lead to wider attention on your crystals?
TT: We published some papers about , and on our lab homepage, we had a picture of our crystals. They weren鈥檛 large, maybe 1 millimetre in length, but that鈥檚 how other people found out we were making these crystals. If Kenji had not discovered this [luminescent] aspect of hexagonal boron nitride, maybe none of this would have happened.
At that point, graphene had recently been discovered. Why was this so important to your work?
TT: Graphene is a single-layered material. Because of this, the big question at the time was what could we put it on to study it? This was a big issue for the 2D materials community because an atomic monolayer like graphene is very sensitive. If you put it on something that is not smooth, it is easily affected.
Hexagonal boron nitride is a wide bandgap material, so you can put graphene on it and it will not be affected, since the surface of the crystal is atomically flat and there is no electrical leakage. It was already a well-known material, but high-quality crystals were hard to come by.
After we had published our results on hexagonal boron nitride, , a postdoc at the time at Columbia University in New York, contacted me. He asked us to supply one of our crystals because he wanted to measure a mechanical property of graphene on it.
What happened next?
TT: At that time, no one knew whether it was possible to combine two kinds of atomic layer. Lee published his report. And then, a year or so later, there were two other young guys at Columbia who wanted to use our crystals as a substrate, so we sent crystals to then PhD student 鈥� now a professor with the University of California, Santa Barbara 鈥� and 鈥� then a staff postdoc and now professor at Columbia University. When they placed graphene on the crystals, it looked very comfortable, there were no dangling bonds, atoms without a bonding partner, which would have an electrical and chemical effect on another adjacent crystal. It was amazing. They reported this at a conference in 2010. That was the break.
Were you surprised by the demand?
TT: As a crystal grower, I was happy to learn that our crystals were somehow useful for the scientific community, though I could not imagine the demand we received. Our crystals were not large, still only 1 millimetre or so 鈥� they looked almost rubbish compared to other large crystals. But people working in 2D materials figured out how to work with them and the field grew up based on studies of materials with similar dimensions.
Have the crystals you grow changed over the years?
TT: The size is the same, there hasn鈥檛 been much progress from my side in that respect. Even 20 years since we first made the crystals, it is still just 1 millimetre. But, for instance, initially our crystals had, say, 10 parts per million levels for carbon [impurities], now it is 10 parts per billion.
Is it difficult deciding who to send crystals to or are you able to say yes to everyone?
TT: I try to say yes to everyone. One important thing is that our institution is a government learning institution, so our export control office makes some careful judgements with respect to potential collaborators.
Do you think the demand for these materials will continue?
TT: At the beginning of 2010, when people started asking for our crystals, Andrea Young said, 鈥淵ou鈥檝e become a very busy guy!鈥� I supposed that might go on for two or three years and then calm down, but now it has been 10 or 15 years.
Year by year, there are new findings in graphene research. Things like twistronics, the idea that if you layer two sheets of graphene on top of each other with a twist in the orientation, new properties appear. The activity in the field is amazing. But how much longer? I am not in a position to say.
I think the idea of using hexagonal boron nitride crystal for 2D material systems will continue, but not only with our crystals. There are some younger researchers looking at new ways of growing hexagonal boron nitride crystals, with a method called large area deposition. I look forward to their success.
What is that technique?
KW: Large area deposition is where atoms are deposited on a surface from a vapour or liquid, for instance by chemical reactions. It is extremely difficult to obtain hexagonal boron nitride from large area deposition with a similar high quality to that of crystals grown under high pressure.
TT: There are already some papers on growing hexagonal boron nitride in this way, but we have to wait, we shouldn鈥檛 rush them. I believe, in time, they will be successful.
Are you teaching younger researchers your techniques?
TT: One of the problems with our institution is age: the young people here are still over 40 years old. I would like to have some young people join. The number of students taking a doctoral course is getting smaller and I am quite worried about that.
When you both retire, what will happen to the field if nobody takes over?
TT: We have been open about how to make the barium solvent we use to create the high-quality crystals, it鈥檚 not a secret. It鈥檚 all in the open, either through a patent or a journal paper. I don鈥檛 know why there are no companies making them yet. Probably, it wouldn鈥檛 make any money!
