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

Annular electrodes as PCM solution (Part 1)

Posted: 24 May 2013 ?? ?Print Version ?Bookmark and Share

Keywords:phase change memory? annular electrode? lithography?

Once the IMH forms, there is a discontinuity between the electrical conductivity of the crystalline material and the higher electrical conductivity of the molten material. This discontinuity acts to localise the current. At this point, one of two things can occur: the hot spot can increase in temperature for constant size, or the hotspot can increase in size for constant temperature. If it is energetically more efficient to raise the temperature of a thin surface layer around the hot spot by some amount T, up to Tm, the hot spot will expand at a constant temperature close to the melting temperature. As long as the volume of the surface layer is less than the volume of the expanding molten hotspot, the rate of expansion will be determined by the reset current. For this to succeed, the localized reset current must supply sufficient energy to raise the temperature of the thin surface layer by T and provide for the latent heat of melting.

As the total volume of molten material increases, the growth rate will decrease. For a large-diameter electrode, the reset current density required to form an IMH in the central region of the electrode must compensate for heat loss from the edges of the electrode. In the case of a smaller-diameter device such as that shown in figure 4(b), the same current density will be able to offset the edge losses. In that situation, if sufficient additional surface area is available to allow for the formation of a 3-nm- to 5-nm-diameter molten hotspot, the electrode contact current density Jc will remain approximately the same, resulting in a reduction in the reset current.

If the IMH model correctly describes the situation, then it would appear that an annulus thickness t of about 6 to 7 nm represents a limit. As figure 3 shows, we have to limit our device to about 40 nm in order to obtain a 50% reduction in Jc; once the edges become significant, we need to make an increase in the value of Jc. Figure 4(c) shows an annular electrode device where Jc will remain the same as for figure 4 (a) and (b), assuming t is sufficient to provide the area for the hot spot to form.

Both solid and annular electrodes will be operating in the quasi-constant current region illustrated in figure 2. In the case of solid electrode devices, once the diameter of the device is reduced to the point at which the edge losses become significant, the current density must be increased to allow the creation of the molten hot spot; this is demonstrated by the steep part of the current density curve in figure 2 for sub 20-nm diameter solid-electrode surface devices. For annular devices, depending on the diameter, the Jb = Jc(ac/Ab) relationship may no longer hold. The value of Jc may be much higher than used to obtain the representative characteristics of Jb in figure 2.

Are annular electrodes practical?
All is not lost, and for the optimists looking towards sub-20-nm PCM scaling, the annular electrode PCM structure may have something additional to offer. Certainly, it is the reason for renewed interest in the annular electrode structures. As discussed earlier, the technique may still provide a means of reducing the electrode body current density Jb imposed on the memory matrix isolation devicewith one important extra consideration.

Thermal coupling between devices in PCM memory matrices at sub-20-nm lithographic dimensions is going to present a real problem and it will remain a problem. Annular electrode PCM device designers may be able to use the effect to advantage and negate the pessimistic predictions of current density that arise from figure 2 for sub-20-nm devices employing solid electrodes. When the inside edges of the annulus are in the thermal-coupling range (less than 20 nm), it should be possible to create the situation that exists in a large-area device that allows for the formation of the IMH (figure 5). According to the model, when the annulus is very thin, the current density at the electrode-active interface would need to increase to deal with the thermal losses from both edges. With thermal coupling and a longer reset pulse, it should be possible to create a thermal situation in which a hot spot can form as a result of a reset current density much less than would be necessary if thermal coupling did not occur. Figure 5(d) illustrates the example where thermal coupling is not a contributing factor; in this case if the annulus is wide enough the device will act as a normal annular electrode device. If the edge contact is very thin, Jc will need to be increased.

Figure 5: The distance between the inside edges of a sub-20-nm annulus minimises current density Jc (a through c). When the inside edges of the annulus are separated by greater than 20 nm (d), Jc will be a strong function of annulus thickness t when t is less than 7 nm.

?First Page?Previous Page 1???2???3???4???5?Next Page?Last Page

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