*2700th posting: new generation of hybird memristive nanodevices and an update of HP labs and its memristive products

Hard to believe this is the *2700th posting but yay! To commemorate this special occasion I’m featuring two items about memristors, work on protein-based memristors and an update of my Feb. 7, 2013 posting on the HP Labs and its promises of memristor-based products.

Michael Berger’s Dec. 16, 2013 issue of Nanowerk Spotlight focused on memristor research from bioengineers at Singapore’s Nanyang Technological University (Note: Links have been removed),

 Based on the rapid development of synthetic chemistry and bioengineering, researchers have begun to build hybrid nanostructures with various biomolecules to fulfill the functional requirements of advanced nanocircuits. Proteins already perform functions such as signalling, charge transport or storage, in all biochemical processes.

“Although the diversity of these natural molecules is vast – for instance, more than a million variants of an individual protein may be created via genetic engineering – tailoring their structures to fit the variable and complex requirements of both the biological and non-biological world is achievable by leveraging on the rapidly developing bioengineering field,” Xiaodong Chen, an Associate Professor in the School of Materials Science & Engineering at Nanyang Technological University, tells Nanowerk. “On a parallel note, bioengineering may provide an alternative approach to tune the structural and electronic properties of functional molecules leading to further development in the field of molecular electronics.”

Berger provides more context on this work by way of a 2011 Spotlight about the research (featured in my Sept. 19, 2011 posting) and then describes Chen’s latest work,

In new work, reported in a recent edition of Small (“Bioengineered Tunable Memristor Based on Protein Nanocage”) Chen and his team demonstrate a strategy for the fabrication of memristive nanodevices with stable and tunable performance by assembling ferritin monolayer inside a on-wire lithography-generated ∼12 nm gap.

Whereas the protein-based memristor devices in the previous work were fabricated from the commercial horse spleen ferritin, the new work uses the unique high iron loading capacity of Archaeoglobus fulgidus ferritin (AfFtn).

“We hypothesized that if the composition of this iron complex core can be modulated, the switching performance of the protein-based device can be controlled accordingly,” says Chen.

They found that the (tunable) iron loading in the AfFtn nanocages drastically impacts the performance of the memristive devices. The higher iron loading amount contributes to better memristive performance due to higher electrochemical activity of the ferric complex core.

This work is not going to be found in any applications for molecular devices at any time soon but it seems promising at this stage. For those who’d like more information, there’s Berger’s article or this link and a citation to the researchers’ paper,

Bioengineered Tunable Memristor Based on Protein Nanocage by Fanben Meng, Barindra Sana, Yuangang Li, Yuanjun Liu, Sierin Lim, & Xiaodong Chen. Article first published online: 19 AUG 2013 DOI: 10.1002/smll.201300810
© 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

This paper is behind a paywall but Wiley does offer a number of viewing options at different price points.

HP Labs and its memristor-based products

Following on last year’s Feb. 7, 2013 update (scroll down about 1/2 way), it seems like another annual update is in order unfortunately, the news seems like a retread. Memristor’-based devices from HP Labs will not be launched (in the marketplace or even to show at technology shows) this year either. In fact, any sort of launch is much further in the future according to Chris Mellor’s Nov. 1, 2013 article for The Register; Note: Links have been removed),

HP has warned El Reg not to get its hopes up too high after the tech titan’s CTO Martin Fink suggested StoreServ arrays could be packed with 100TB Memristor drives come 2018.

In five years, according to Fink, DRAM and NAND scaling will hit a wall, limiting the maximum capacity of the technologies: process shrinks will come to a shuddering halt when the memories’ reliability drops off a cliff as a side effect of reducing the size of electronics on the silicon dies.

The HP answer to this scaling wall is Memristor, its flavour of resistive RAM technology that is supposed to have DRAM-like speed and better-than-NAND storage density. Fink claimed at an HP Discover event in Las Vegas that Memristor devices will be ready by the time flash NAND hits its limit in five years. He also showed off a Memristor wafer, adding that it could have a 1.5PB capacity by the end of the decade.

Fink spoke about the tech in June, but this week a HP spokesperson clarified to The Reg:

As with many other ground-breaking technologies being developed at HP Labs, HP has not yet committed to a specific product roadmap for Memristor-based products. HP does have internal milestones that are subject to change, depending on shifting market, technology and business conditions.

Every time I read about it HP Labs’ memristor-based products  they keep receding further into the future. Compare this latest announcement with what was being said at the time of my Feb.7, 2013 posting,

… Stanley Williams’ presence in the video reminded me of the memristor and an announcement (mentioned in my April 19, 2012 posting) that HP Labs would be rolling out some memristor-enabled products in 2013. Sadly, later in the year I missed this announcement, from a July 9, 2012 posting by Chris Mellor for TheRegister.co.uk,

Previously he (Stanley Williams) has said that HP and fab partner Hynix would launch a memristor product in the summer of 2013. At the Kavli do [Kavli Foundation Roundtable, June 2012], Williams said: “In terms of commercialisation, we’ll have something technologically viable by the end of next year [2014].”

To be fair, it seems HP Labs had abandoned plans for a commercial launch of memristor-based products even in 2013 but now it seems there is no roadmap of any kind.

* Corrected from ‘3000’ to ‘2700’.

Leave a Reply

Your email address will not be published. Required fields are marked *