Stay Informed:

COVID-19 (coronavirus) information
Zoom Links: Zoom Help | Teaching with Zoom | Zoom Quick Guide

ECE Seminar Series: Thermoelectricity at Nanometer Scale & Phase Change Memory

Speaker Name: 
Ali Gokirmak
Speaker Title: 
Speaker Organization: 
University of Connecticut
Start Time: 
Monday, November 18, 2019 - 10:40am
End Time: 
Monday, November 18, 2019 - 11:40am


Energy exchanges associated with generation and recombination of free charge carriers give rise to dramatic effects in small scale structures experiencing high electric fields (~30 MV/m), high current densities (~107 A/cm2), high temperatures (~1000 K – 2000 K) and strong thermal gradients (1 - 50 K / nm). Breaking of the local equilibrium of generation-recombination balance by an external field gives rise to a heat flux in the direction of the minority carriers. At temperatures close to melting, this effect overcomes the effect of electronic convective heat flow by majority carriers moving in the opposite direction. We have directly observed this Generation-Transport-Recombination (GTR) component Thomson Heat (thermoelectric heat flow in uniform materials) in lithographically defined silicon microwires. 

These rather extreme electrical and thermal conditions come into existence in write/erase operations of phase change memory (PCM) devices, which became viable non-volatile data storage devices with the ability to manufacture devices with < 50 nm critical features.  Phase change memory (PCM) is a high-speed resistive non-volatile memory (RRAM) technology that utilizes the resistivity contrast of amorphous and crystalline phases of glassy materials to store information. PCM devices can be repeatedly changed from one phase to the other by local heating and melting using electric current. PCM has the prospect to be integrated on top of VLSI circuits, producing computer chips with large volume (> 250 GB) of non-volatile memory in a single chip. This embedded storage can eliminate the need for additional computer memory (DRAM) or a motherboard, speeding up computer performance by > 1000x for data intensive applications.



Ali Gokirmak received his BS Degree in Electrical Engineering and Physics from the University of Maryland at College Park in 1998 and received his PhD in Electrical and Computer Engineering from Cornell University in 2005. He has served as a Postdoctoral Research Associate at Cornell for one year in the same group. Ali has been a Professor in Electrical & Computer Engineering at the University of Connecticut since 2006.


Event Type: