Straining for quicker computers - accurate nanoscale measurement project
The European Metrology Research Programme’s Nanostrain project could deliver faster computing in a less power-intensive way. Eoin Redahan asked Dr Carlo Vecchini, one of the project leaders, about the initiative.
How will you make computer processing speeds faster while using less power?
The Nanostrain project will lay the fundamentals for a new paradigm in computing technology, where electromechanical coupling will replace charge transport. In this project, the technology necessary to precisely and accurately control strain at the nanoscale will be developed. This will represent a crucial step in the development of new devices based on the control of strain, such as the piezoelectric effect transistors (PET) proposed by IBM. Recent calculations based on this model found devices could be made around 10 times faster and consume 100 times less power than traditional transistors. With this technology, piezoelectric materials are driven to expand and shrink while being constrained by a solid yoke. The process creates strain that can be used to control the flow of information inside a PET. This will create a completely new generation of central processing units that will finally be able to break the 3GHz limit current technology has reached.
How will the project work?
The main objective of the project is to create and develop the advanced measurement capabilities necessary to precisely characterise strain in nanomaterials. The UK’s National Physical Laboratory has assembled world-leading experts in the relevant scientific areas. This comprises several European national laboratories and a consortium of collaborators, including world-class research instrument facilities at the European Synchrotron Radiation Facility, in Grenoble, France, and nine commercial companies spanning a wide range of applications. Regular meetings are organised among the project partners and collaborators to facilitate knowledge transfer and disseminate achievements.
What is the importance of being able to deliver highly accurate measurements of strain?
There is currently a lack of advanced measurement systems for strain at the nanoscale, and this project aims to fill the gap. Reliable, accurate and traceable measurement of strain in functional materials under industrially relevant conditions of high stress, electric field and at nanoscale dimensions are essential in providing the data to support emerging technologies.
What type of piezoelectric materials do you intend to use?
We will use materials that exhibit very high piezoelectric coefficients. New generations of piezoelectric materials are formulated to exhibit superior electro-mechanical properties, which are particularly appealing for technological applications. They can be prepared in many different forms, such as thin films and single crystals of various dimensions, which can show very large piezoelectric coefficients.
How would a piezoelectric effect transistor be better than existing technology?
PETs have the potential to be orders of magnitude better than existing technology, because they circumvent the speed/power bottlenecks of current CMOS technology. A jump of one order of magnitude in speed and two orders of magnitude less in power consumption represent a complete game change. The Nanostrain project only started on 1 July 2013 but advances in several areas have already been reported.
How soon will the public see the benefits?
The PET technology is estimated to take up to 15–20 years to become commercially available, because it is based on a completely new paradigm. Areas that will see benefits from this work include medical applications with smaller and faster devices, telecommunications for faster data transport, aerospace and automotive with reduced power consumption, and new storage methods.
The Nanostrain project, which was announced on 4 November 2013, will run for three years. For a full list of the organisations involved and more information on the project, visit www.piezoinstitute.com/resources/emrp_nanostrain.php