A revolution of nanoscale dimensions

Revolutions in technology can be disruptive to society, but they also propel society forward to change. The industrial revolution of the eighteenth century led to the end of serfdom in Russia and stimulated the start of industry along with the emigration of Europeans to the ‘New World’ in search of opportunity. Societal revolutions in Europe thereby generated a workforce and a need to educate this workforce. The first societal revolution that influenced me personally was Sputnik in 1957, which resulted in a rapid increase in US government support for science and technology research, and for education to enable an increasing workforce to catch up with Russian achievements. At that time, many young people studied science and engineering, and made their careers by bringing ideas to the industrial sector through start-up companies helped by federal matching funds and private equity.

In 1967, I had the opportunity to teach solid-state physics to engineers at the Massachusetts Institute of Technology (MIT), because the Dean of Engineering, Gordon Brown, who was greatly influenced by his experience during the Second World War, believed that solid-state electronics was rapidly growing in importance. Brown believed that silicon-based electronics would provide new levels of functionality and efficiency to electronic devices, first for the space programme and then for society more broadly. He and MIT President Jerome Wiesner were both influential advisors to the US government on the development of technical capacity in the country as a means to stimulate US industry and thereby to help Europe and Asia rebuild their economies after the devastation of the Second World War.

Between 1960 and 1990, the US government slowly started stimulating the development of solid-state electronics ending with the National Nanotechnology Initiative. Although earlier efforts had begun in powerful industrial laboratories, such as Bell Labs, IBM, Belcore and General Electric Research Laboratories, these laboratories were looking to the universities for young, talented scientists and engineers. As a consequence, the government's support for the education and training of scientists and engineers increased dramatically in response to these employment opportunities. The demand for graduates increased as the private sector in Silicon Valley grew, and smaller technology centres flourished all over the United States. Regional centres for nanoscience and technology also developed — based on local government, federal government and industrial support. There was more demand for higher education in scientific disciplines, and immigration of young, talented people to the United States greatly increased; from Europe at first, and then from South America and Asia.

Much of this activity was stimulated by efforts to maintain the forward movement of Moore's Law — a drive to diminish the size of electronic device features, but to increase functionality and decrease the cost of computational operations. The consequences of these functionality increases and cost decreases fuelled the development of computer control of industrial operations and the growth of electronics, communications, automation and information storage. This revolution affected everyday life at home, including children's toys and childhood behaviour in preferring computer-based toys to outdoor play and exercise.

Society now expects such revolutions — with the aid of electronics and nanoelectronics — to continue into the future. The assumption is that discovery in science and technology will not only be maintained but will also continue to grow at an accelerated rate. Some engineers and scientists feel that continuing in our present mode of operation is unrealistic. With length scales of device features reaching the 10-nanometre scale, present device and research strategies based on silicon will soon be reaching their natural limits, and single-molecule electronics will take over. Likewise, computer scientists have been developing ever-more efficient software at a pace to match the decreasing feature size of electronic devices.

“The timing now seems right for the next revolution”

Computer science experts are now asking what will happen when the Moore's Law paradigm slows down. The training of computer scientists has been carried out with reference to hardware based primarily on silicon technology; hence, this results in a gap in thinking between those making the devices and the computer giants. It is obvious that computer scientists will become more integrated with interdisciplinary teams of materials scientists and engineers, working together in university communities and with complementary partners in industry and national laboratories. All of the scientists in these assembled collaborations will bring diverse capabilities, practical instrumentation and expertise. Young students will benefit from this environment supported by federal funding initiatives. Moreover, I see such programmes developing worldwide, perhaps stimulated by international professional societies. For example, Europe has the Future & Emerging Technologies Flagship Initiatives, and the Americas and Asia also have initiatives, but these programmes are less structuredat present.

The scenario described below may provide the next revolution in materials and, from my perspective, could constitute the hardware of future electronic and computational devices. The tools currently under development by researchers focus on new materials bringing new capabilities to electronics. Research developments in electronic devices have, for the past decade, focused on 2D layered materials, such as graphene, as these materials produce functionalities that are not available with silicon technology, which has been the main focus of Moore's law. Multiple graphene layers can be stacked with the same Bernal stacking order as in graphite, or they can be stacked with different twist angles between adjacent layers with respect to their crystalline axes, which allows some control of their physical properties.

But this is not all that nanoscience can do with carbon nanostructures. Long graphene ribbons with narrow widths exhibit a higher carrier density along one edge configuration relative to another (for example, at zigzag edges relative to armchair edges). This difference in the edge structure can, for example, be exploited in plasmonic devices, in which the carrier densities can be controlled. In the same vein, the different magnetic properties produced at the various graphene edge configurations can be exploited in magnetic devices.

Beyond electronics-inspired applications, and as a consequence of its simplicity together with the extensive study it has recieved, graphene has become a standard reference material for nanoscience. In contrast to monolayer graphene, with its high in-plane electrical conductivity, the inter-planar electrical conductivity is very poor for bilayer graphene and few-layer graphene. The difference in electronic properties between the different symmetries of monolayer relative to bilayer, few-layer, and many-layer graphene are important for its adoption as a standard in nanoscale metrology.

At present, other few-layered materials are also being introduced for electronics applications. As a consequence of the high in-plane electric conductivity of monolayer graphene, significant attention has been given to semiconducting hexagonal boron nitride (h-BN) because of its large energy band gap (∼5 eV) and very flat surface in crystalline form. Moreover, h-BN has a similar lattice constant to graphene. For these reasons, h-BN has been used as a common substrate in the research community for other layered materials. Major efforts have been made in the field of few-layer transition metal dichalcogenide (TMD) materials because of the large number of available transition metal and chalcogen species. Tailored TMD materials with a large variety of electrical and thermal properties — for a range of applications — can be prepared depending on the specific chemical species and mixtures of chemical species selected for the cations and anions for negatively and positively charged building blocks. Further choices in compositional combinations for tailoring electronic, optical and thermal properties can be made when including lower-symmetry layered materials such as phosphorene (a single layer of phosphorus) and other in-plane anisotropic materials within the category of layered materials.

Clearly, the applications of such layered nanosheets to electronics and computer science are at an even earlier stage than for graphene-based electronics, and the impact of all of these few-layered thin materials on the successor to Moore's Law is not yet known. It is likely that many niche applications will be found, and successful start-up companies may soon have a large impact on future electronic, optical and mechanical devices as well as on computer science.

The up-and-coming, educated workforce is highly excited about future possibilities for nanoscience and nanotechnology, and I am personally also thrilled about this progress because of my past experiences in the dramatic changes that occurred as a result of the two earlier revolutions described above. The timing now seems right for the next revolution. Many ideas are surfacing in the research community and, more widely, an expanding, experienced and knowledgeable workforce has assembled. Governmental and private funding agencies have established collaborations across academia, industry and national laboratories, and in many cases the collaborations are international, to the benefit of society.

Affiliations

1. 
   

   Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, USA.

   
   
   
   
   • Mildred S. Dresselhaus