News

doi:10.1038/nindia.2013.9 Published online 24 January 2013

Molecule bed for future memory chips

Subhra Priyadarshini

With electronic devices getting smaller and thinner, further miniaturising the dimensions of memory chips is a challenge for researchers. An international team of scientists, including some from India, has now created a magnetic supramolecule — by chemically binding two molecules — to make a device that can transfer and store information in a single molecular magnetic layer. This, they say, could help make scalable, non-volatile molecular memory devices that can work at room temperature1.

It will be possible to store information in a single molecular magnetic layer.
© MIT

The researchers from Indian Institute of Science Education and Research (IISE)-Kolkata and Massachusetts Institute of Technology (MIT) along with collaborators from Peter Grünberg Institute in Jülich and Gottingen University, Germany have demonstrated a new phenomenon taking place at the molecular level, providing a bottom-up approach in building electronic devices of the future.

"To build future electronic devices we need to work with individual atoms and molecules and use them as building blocks to construct devices for storage, sensing and processing information. Molecules are one such ideal candidate," says Karthik V. Raman, one of the researchers from MIT, currently a visiting scientist at IBM's India Research Labs.

The researchers worked on the knowledge that molecular properties could be engineered using chemistry. They picked up fragments of graphene (which they call phenalenyl) and utilised each of its molecules as a storage element or a 'bit where the binary states of '0' or '1' are represented by the electron spin state in the molecule – 'spin-up' or 'spin down'.

"Phenalenyl are an exciting class of molecules. There is a lot of scope in modifying the structure and spin chemistry of phenalenyl based molecules which can be precisely engineered by using chemical synthesis," says Swadhin Mandal the lead chemist from IISER-Kolkata.

"Since the size of molecules is about a nanometre, this lets you have high storage density of about 1000 TeraByte/sq inch – this is 1000 times denser than what is available in market as hard drives," says Jagadeesh Moodera, a senior research scientist and project leader from MIT.

The researchers encountered a problem of stabilizing and electronically accessing the spin state of the molecule at higher temperature close to room temperature. They had to think of a clever method to do so. They tried to grow these molecules on a clean cobalt surface in a high vacuum chamber. The molecules bound (bind?) to cobalt atoms. Due to the interface spin chemistry, the magnetic properties of the molecule were stable even when close to room temperature.

Top:(L to R) The IISER team Tamal Sen, Swadhin Mandal & Arup Mukherjee with a model of the supramolecule. Bottom: MIT reseracher Karthik Raman (left) and MIT project leader Jagadeesh Moodera.
@ IISER & MIT

"We found that magnetization of the molecule can be switched between the two states by means of external magnetic field. Electronically, this translated to a large change in the resistance of the device," Raman adds.

The team formed a magnetic supramolecule at the interface with the cobalt surface, that gave rise to a "new, interesting" effect in the device. "We saw that the supramolecule has the capability to filter out certain electrons. This means only electrons with a particular spin can pass through whereas other electrons faced a barrier. Such a phenomenon exhibits an interface magneto-resistance effect," he explains.

The researchers say this molecular 'bit' doesn't just store information but also serves as a sensor. It is an integrated storage-sensing system different from existing technology in hard drives where the sensor element is attached to a flying read head and is physically separated from the storage media. "This new design greatly simplifies the challenges of fabricating a high density storage device," Raman adds.

However, some challenges in fabrication of such molecular devices will have to be overcome in times to come. "We are confident that in next 5 years significant progress will happen in this area, taking us towards molecular-level electronics and quantum computing devices," Raman says.


References

  1. Raman, K. V. et. al. Interface engineered templates for molecular spin memory devices. Nature 493, 509-513 (2013)  | Article |