Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Memory/Storage

Phase change memory advances: Threshold switching

Posted: 07 Mar 2016 ?? ?Print Version ?Bookmark and Share

Keywords:conduction? non-volatile memory? phase change memory? PCM? GeTe?

To temper our enthusiasm on the subject of the possibility of a more powerful feedback mechanism resulting from a conductivity discontinuity Manuel Le Gallo offered the following by way of a suggestion: "However we do not think that there is a significant discontinuity in conductivity between the amorphous and molten states if those were at the same temperature. To make it clear, the conductivity of the amorphous state at the melting temperature (~900K) is expected to be very similar to the conductivity of the molten state."

Perfection, melting and oscillators
Although in the world of steady state simulation perfection a hot torus would develop, as shown in figure 1(b) from the IBM work, in reality with physical imperfections it is more likely that a localized hot would develop and then spread around and across the bottom electrode surface. Even without any significant local capacitance the same input power, melting might be a possibility in the case of the 15 to 22uW of figure 1(b). Any local capacitance making an additional i = CdV/dt contribution to the current.

As illustrated in Figure 3 there are three possible models for pre-threshold and threshold switching sequence for different device structures. One for the spherical and another for the planar device structures and a third based on possible geometric imperfection of the spherical device.

Figure 3: Three options for an initiating hotspot (IHS) (a) the torus for the "dome" structure, (b) the planar structure and (c) the bottom electrode edge effect. Red colour molten material.

The intriguing question is when does the first molten hot spot occur, is it before or after the maximum voltage has been reached. In the figure 3(a) for illustrative purposes the orange coloured regions are hot amorphous material and the red coloured molten material. In each case the final volume of molten material will be determined by the set current.

In the example of figure 3(b) the single initiating hotspot (IHS) would reach the same temperature and more quickly for the same input power than figure 3(a) where a complete torus would need to reach the same temperature. Figure 3(c) illustrates a third possible example for a spherical or "dome" structure where some form of geometric imperfection occurs. The imperfection could be in the central positioning of the dome relative to the bottom electrode or an irregular imperfection in the form of the bottom electrode Figure 3(c).

The latest IBM results would also account for the reason why we do not see high frequency or for that matter any oscillators based on the use of the negative resistance in threshold switching materials, however careful the design to avoid damaging transients. Without careful design PCM devices in a free running oscillator the build up of temperature might lead to a situation where the material would crystallise. Even in those cases where short lived oscillator performance is obtained frequencies are always limited to values less than the reciprocal of the thermal time constant. Any oscillator frequency would be limited to the reciprocal of the sum of the delay time, plus the transient time plus the recovery time. The lack of any free running oscillators is an observation that was always ignored by the advocates of electronic switching mechanism for memory or threshold switches.

Given the correctness of this latest IBM thermal interpretation of threshold switching and if it is also applicable to memory and threshold switch matrix isolation devices then thermal considerations for array operation as well as cell structure will need to be a major consideration for those planning to build 3D stacked or cross point or 3DXPoint memory arrays be they chalcogenide or oxide based.

[1] Le Gallo, Manuel, et al. "Evidence for thermally-assisted threshold switching behaviour in nanoscale phase-change memory cells"
[2] Neale, Ron. "New Insights on Non-Volatile Memory Conduction", EE Times, published 12/7/2015.
[3] "Physics of Switching and Memory Effects in Chalcogenides K. D. Tsendin" Universal Journal of Physics and Application 2(2): 53-59, 2014 DOI: 10.13189/ujpa.2014.020201
[4] Neale, Ron. "IBM Says PCM Non-Volatility Not Essential", EE Times, published 12/17/2014.
[5] Neale, Ron. "IBM Takes A Second Turn at PCM Drift", EE Times, published 7.31.2015.

About the author
Ron Neale is an independent electrical/electronic manufacturing professional.

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

Article Comments - Phase change memory advances: Thresh...
*? 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