Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Manufacturing/Packaging

Annular electrodes as PCM solution (Part 2)

Posted: 28 May 2013 ?? ?Print Version ?Bookmark and Share

Keywords:self-assembly? sub-lithographic? fabrication process?

The two centre diagrams of figure 4 compare a simple annular electrode structure with the equivalent eight-disc solution for a 50% fill factor. With the same fill factor, the electrode surface area is the same and the Jc/Jb ratio will be the same. Recall, though, that both simulations and experimental results showed lower current density Jb from the self-assembled device than could be explained by dimensions alone. The question that must be answered is why there might be a difference in current densities Jc, Jb and the reset current for the S-A device and a similar diameter annular electrode PCM device when the volume of active material above the electrode is the same.

Figure 4: Comparison of 50% fill factor of eight balls (middle top) with equivalent-fill-factor simple annular electrode.

Reviewing the results
Although there have been simulations, there are no published results for actual sub-20-nm PCM devices using 3-nm-wide annular electrodes or for equivalent S-A fabricated devices, as far as I am aware, to allow for direct comparisons. A similar approach to that used earlier for the annular electrode in part 1 might provide a way of exploring the differences and similarities. In figure 7 of Part 1, a single device was extracted from the annulus and its current density used to estimate the current density for that same device operating embedded in the annulus where the heat losses would be much reduced. If we use that same approach is used for the core annulus device (green) in figure 10 then the situation is as shown in figure 5. For the example using 5-nm oxide disks and a roughly 50% fill factor the extracted irregularly shaped device would have an average surface diameter of between 3 and 4 nm, very similar to the earlier example. Extracting the current density from figure 2 again gives a value of about 2 108 A/cm2.

Figure 5: For the self-assembled structure (left), heat loss is limited along the three electrode paths and by the very close thermal coupling across the oxide. In comparison, and extracted stand-alone device (right) has he lost in all directions.

During reset, the embedded device will not suffer heat loss along the three dissipating electrode surfaces and the very strong thermal coupling across the oxide from the extended electrode surfaces will limit heat loss across oxide disks.

There will, of course, be heat loss in the direction normal to the surface, I have considered that to be the same in the other examples used in this paper. While we cited a two-fold reduction in current density in the earlier example for the annular electrode structure, a much larger reduction factor can be reasonably ascribed in this case for the embedded device vis-a-vis the extracted device. A factor of greater than four would result in reset current densities Jc close to 5 107 A/cm2 with a value of Jb somewhere in the range of 2.5 107 A/cm2. Given that this estimate is valid it does not suggest, even given the ability for S-A to form 5-nm disks, that S-A devices will offer a safe solution to the problem of current density for sub-20-nm devices.

In this paper I have used the term thermal coupling or thermal cross talk as the means by which one region of a device structure communicates with another. For PCM device structures with annular shaped electrodes, the reset current flows around and must heat an electrically passive volume of dielectric. For a complete reset it is necessary to raise the temperature of the surface of the dielectric to the melting temperature of the crystallized active material. This will be a time dependent function and will become more efficient if the passive parts of the structure are smaller and closer together as is the case in the S-A structures. The inverted S-A structure does remove some of the thermal coupling or heat sinking of the active region of the PCM cell to the substrate and may account for some of the reduction in reset current reported for the inverted structure.

Although space does not allow a discussion of possible fabrication problems from inverting the PCM device structure for self-assembly; that is moving the addition of a chalcogenide active memory material like germanium-antimony-tellurium (GST) from a BEOL process, I would like to suggest an approach that would give rise to the same result and possibly offer an alternative means of creating sub-5-nm regions of oxide or dielectric on the electrode surface to achieve the same effect as the oxide islands. The methodology would involve normal processing with the difference of using a BE material that would crystallise in two phases of mixed crystals. It would require the additional property that one crystal phase could be selectively oxidized to create the required surface structure. The deposition of the GST and top electrode would then be added in the normal manner. In the limit, it might even possible to even use a graphene film placed on the electrode surface and sputter a dielectric over it to produce the dielectric islands, then make use of the electrical conductivity of graphene to make a lateral connection to the active material.

The analysis I have presented here does not provide any measure of optimism for the future of sub-20-nm PCM devices using either the annular ring electrode or the techniques of self-assembly. For the near future it would appear that attempts at providing high density PCM will need to rely on the use of MLC-PCM devices in MCP packaging. New results for the MLC-PCMs, will be presented at the upcoming 2013 IEEE International Memory Workshop (IMW), scheduled (May 26-29; Monterey, CA). It would appear that IBM now has answers to the outstanding questions of elevated temperature data retention with write/erase lifetime for MLC PCMs.

Something for PCM watchers to look forward to.

1. K. Lewotsky, "Tech Beat: Nano-structures promise power-sipping PCM," Memory Designline, March 2013.
2. W.I. Park, B.K. You, et al., "Self-Assembled Incorporation of Modulated Block Copolymer Nanostructures in Phase-Change Memory for Switching Power Reduction," ACS Nano 7[3], pp 2651C2658 (2013).

About the author
Ron Neale is the former editor-in-chief of Electronic Engineering. Also, he is the co-author of "Nonvolatile and reprogrammable, the read-mostly memory is here," by R.G.Neale, D.L.Nelson and Gordon E. Moore, Electronics, pp56-60, Sept. 28, 1970.

To download the PDF version of this article, click here.

?First Page?Previous Page 1???2???3

Article Comments - Annular electrodes as PCM solution (...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top