Is Germanium the Next Little Thing?
[heading size=”1″ color=”#fff000″]Breakthroughs in Physics[/heading]
[message_box type=”info/warning” align=”left” color=”#ffffff” background=”#dba044″ border_color=”#201e1e” box_shadow_color=”#201e1e”] Maria Gherasimova is testing the properties of germanium, which could lead to the next generation of transistors.[/message_box]
[stag_intro][stag_dropcap font_size=”100px” style=”squared”]T[/stag_dropcap]he development of a method to mass-produce silicon transistors in the late 1950s followed by the explosion of innovation in Silicon Valley revolutionized the electronics industry and forever changed our lives. While we all know that silicon semiconductor transistors led to smart phones, tablets and a vast array of electronic gadgets, they also led to the creation of synthetic tissue for burn victims, artificial hip joints, deep space exploration vehicles and much more. However, unless you are a physics or chemistry buff, it’s unlikely that you know, or even care to know, the science behind the discoveries.[/stag_intro]
Curiosity of the basic properties and possibilities is what motivates the research of Maria Gherasimova, Ph.D., Assistant Professor of Physics. In particular, Gherasimova is interested in the potential for a novel paradigm of a logical element that departs from the well-established metal oxide semiconductor field effect transistors (MOSFETs) developed in the 1970s that are still the mainstay of silicon chips. While graphene, a homogeneous, two-dimensional, atomically thin, honeycomb-shaped carbon crystal structure, is now the focus of much research in this area, Gherasimova has turned her focus to another member of the periodic table’s carbon family with semiconducting properties: germanium. In fact, researchers first experimented with germanium before settling on silicon transistors.
[one_half]The nature of silicon semiconductor transistors is such that they can “speak” the binary language of the logical operations required for electronics through a specific, albeit extremely small, assemblage, the MOSFET. This grouping of silicon, “doped” (a process in which impurities are intentionally introduced) at the lattice level with other metals and oxides, conducts an electric signal used for binary logic.
The miniature assemblage has been packed into increasingly smaller spaces through improved lithography over the past few decades, allowing for an increase in the number of functions performed per unit time, i.e. faster computation speeds in increasingly smaller products. Still, this innovation trajectory will have limits, since at some point the maximum number of MOSFETs that can be packed onto a small area on a chip and function finally will be reached.
Gherasimova is experimenting with quantum dot fabrication methods to create a new type of medium for binary language, quantum cellular automata (QCA), consisting of germanium island clusters (cells) on a substrate of silicon. The QCA cells, rather than interacting via electric current like MOSFETs, interact via electrostatic fields. A QCA wire would transmit information via changing the state of the neighboring cells that are used to encode the binary “0s” or “1s,” thus achieving a logical circuit for information transmission without the use of electrical current.
Critical factors impact this process. For one, germanium island clusters need to have a consistent distribution and size on the silicon substrate, which is difficult to achieve through self-assembly since germanium islands naturally nucleate irregularly at random locations on a silicon substrate. [/one_half]
[one_half_last][one_half_last][message_box type=”info/warning” align=”left” color=”#dba044″ background=”#201e1e” border_color=”#dba044″ box_shadow_color=”#dba044″]While graphene is now the focus of much research in this area, Gherasimova has turned her focus to another member of the periodic table’s carbon family with semiconducting properties: germanium.[/message_box]
Next, the germanium islands in the clusters need to be sufficiently close (only nanometers apart) for charge tunneling and the clusters need to be close enough for the electrostatic communication to occur (tens of nanometers apart). So Gherasimova is focused on identifying the optimal method to achieve uniform, sufficiently close distribution of these germanium islands.
An institutional seed money grant award in 2012 funded Gherasimova’s multiple trips to IBM’s T. J. Watson Research Center in Yorktown Heights, New York, where she performed experiments using IBM’s modified ultra high vacuum (UHV) Hitachi transmission electron microscope (TEM) that has a focused ion beam (FIB) implantation capability within the same UHV environment. This enabled experimentation and observation of germanium deposition and self-assembly on FIB-modified surfaces within the apparatus. FIB surface treatment on the silicon substrate was used to “herd” germanium dot growth away from untreated surface area to the desired locations, i.e., control the location of the clusters as well as the individual islands within the clusters.
In these recent experiments, Gherasimova focused on utilizing self assembly of small clusters of islands to achieve small inter-island separation distance instead of relying on the deterministic one-to-one ratio between the islands and FIB-encoded sites, which was the goal of her earlier work. Self assembly has the potential to overcome the limitation on minimizing this distance imposed by surface diffusion during the deposition.
Gherasimova is continuing her research on campus by analyzing the data and preparing the results for publication. She has presented her recent findings at the Materials Research Society meeting in the fall of 2012, and Connecticut Microelectronics and Optoelectronics Consortium Symposium in the spring of 2013. [/one_half_last]