back to S1S home

Sunday, July 13, 2014

#296 Alternative PLD techniques

Following on comments in blog #293 let me point out that the technology of thin-film formation using liquid precursors expands rapidly in semiconductor device fabrication. It is no longer primarily photoresist deposition. Consider organic polymer semiconductors,  colloidal solutions containing nanodots, nanowires or nanotubes, low-k interlayer dielectrics, etc.


Responding to those emerging needs the arsenal of physical layer deposition techniques also grows well beyond the best established spin-on deposition. Printing, including ink-jet-like printing and nanoprinting, micro-spray and mist deposition (see next blog for more on this one) are finding their way into semiconductor manufacturing where a rigid, circular wafer is no longer the only type of the substrate (think very large area substrates, flexible sheets and ribbons substrates, roll-to-roll processing, etc.)


Posted by Jerzy Ruzyllo at 01:17 PM | Semiconductors | Link

Sunday, July 6, 2014

#295 July blogs

Due to the extensive travel, July blogs will be delayed, but stay tuned...

Posted by Jerzy Ruzyllo at 01:10 PM | Semiconductors | Link

Sunday, June 29, 2014

#294 Doubts about below 7 nm technology node

In relation to the blog # 291, I am taking note of the growing concerns regarding affordability of the gate length technology below 5 nm. See recent comments along the same lines.

Posted by Jerzy Ruzyllo at 04:00 PM | Semiconductors | Link

Sunday, June 22, 2014

#293 More on PLD

More on PLD because the pool of materials used in semiconductor technology to form thin-film using this deposition technique is growing. In general, process of physical liquid deposition forming thin-film involves application of the liquid precursor to the surface of the substrate followed by thermal curing causing vaporization of the solvent and solidification of the film. In the case of painting (see previous blog) liquid precursor (paint) would be applied using brush. Simple, right?


Well, PLD methods such as painting, brushing, screen-printing, etc. are too crude to be used in semiconductor device manufacturing where requirements regarding control over thickness and uniformity of the film are extremely stringent. The most common PLD process in semiconductor manufacturing is a process of spin-deposition of photoresist which initially in the form of a viscous liquid ends up on the surface of the wafer in the form of highly homogenous thin film.


More on the alternative PLD techniques later...

Posted by Jerzy Ruzyllo at 10:17 AM | Semiconductors | Link

Sunday, June 15, 2014

#292 Physical Liquid Deposition, PLD (think... painting!)

Thin film technology is a foundation of semiconductor manufacturing. In the process of thin film formation the source material (starting material) can be a gas (typical situation in the most of the Chemical Vapor Deposition or CVD processes), a solid (common in Physical Deposition Processes or PVD) or a liquid in which case the term Physical Liquid Deposition, or PLD, seems to adequately describe the nature of the process.


At the first glance the concept of Physical Liquid Deposition may not be all that obvious. In the case you don’t immediately see what PLD is all about think…. painting!


More on PLD next time

Posted by Jerzy Ruzyllo at 12:54 PM | Semiconductors | Link

Sunday, June 1, 2014

#291 "Concorde" and 5 nm technology node - what do they have in common?

You remember "Concorde", a supersonic passenger jetliner capable of moving passengers across the Atlantic in 3 hours of so? It was almost 2 times faster that any other passenger jet. How come then this engineering marvel was retired for good in 2003? Well, it was simply too expensive. In other words, gains resulting from the technological achievement did not justified a very high cost of its implementation.Besides, what if on a given day the time of door-to-door trip between offices in Manhattan and the City of London would be determined by the time of the taxi rides to and from the airports rather than by the time of the flight itself? In this context a supersonic speed of "Concorde" doesn't look so glamorous, right?


I have a feeling that the same will happen to 5 nm and below IC technology nodes. In other words, I venture to predict that 5 nm gate length technology will be never brought to fruition as a commercial, mass-manufactured  product for logic applications because of the cost of its implementation being significantly higher than the potential profit gains resulting from the improved transistor performance. Besides, what if overall circuit performance would be limited by the interconnect and package related delays and dissipated heat management problems and not by the performance of the ultra-short gate length transistor? In this context the fact that the gate of the transistor will be some 10 atoms long would look somewhat less exciting, right?


Well, I hope you see what I am trying to say. But just to make it clear, I fully understand how remote  the parallel between supersonic "Concord" and 5 nm integrated circuit technology node is. At the same time, however, I can not help but to take note of the fact that the times have come when outstanding, ground breaking, progress defining, ready to implement accomplishments of science and engineering are being  shelved because we are not willing to pay for them. To me, it looks like a somewhat new chapter in the history of our technical civilization.
Do you agree with me?

Posted by Jerzy Ruzyllo at 06:16 PM | Semiconductors | Link

Sunday, May 25, 2014

#290 Semiconductors - a truly global affair

Here is an observation coming from an old timer…


Semiconductor science and engineering has become a truly global affair. While it is obviously a status quo now, it was not the case in the past. There were times some 50 years ago when the vast majority of semiconductor related activities, whether in terms of academic research or industrial manufacturing, were taking place in the United States. That is not to say that the technically and scientifically meaningful developments with respect to semiconductor device fundamentals were not simultaneously taking place in Europe and Japan. For instance, important contributions to the development of transistor after World War II came from the Radar Research Establishment in Malvern, Great Britain while other key electronic and photonic device related innovations came at that time from Japan and Russia. Yet, when the 60’s and 70’s rolled in the overwhelming impact which the U.S. had on semiconductor science and engineering worldwide was evident.


Much has changed since those early years. Today, the vast majority of worldwide wafer processing capabilities are installed in Asia with Taiwan leading the field. China is bound to become a major player with wafer processing capabilities expected to double within the next few years while Samsung Electronics is projected (did it already happen?) to take over Intel as the largest semiconductor manufacturer in the world. In short, semiconductor production and R&D activities are gaining momentum all over the world including countries where semiconductor related industrial and educational initiatives are still a relative novelty.


All of those factors clearly illustrate a major paradigm shift in the distribution of semiconductor related activities across the world which took place over the last two decades or so.

Posted by Jerzy Ruzyllo at 08:36 PM | Semiconductors | Link

Thursday, May 15, 2014

#289 Light emission and detection - it is (almost) all semiconductors

The emission and detection of light across the wavelengths range from far infrared to deep UV is these days an integral component of technical infrastructure through which we interact with the world.


Both light emission, based on the conversion of the electrical signal into light and light detection, in which case conversion of light into electricity is taking place, are based almost solely on semiconductor devices. Whether these are for instance light emitting diodes (LEDs) or image sensors such as CCDs (charged coupled devices) both emission and detection of light is based on the unique physical properties of devices constructed using semiconductor materials.


On more than one occasion I was surprised to see that this obvious fact is not all that obvious to some people...

Posted by Jerzy Ruzyllo at 03:05 PM | Semiconductors | Link

Sunday, May 11, 2014

#288 VeSFET: an alternative transistor architecture

Already on few earlier accassions I was bringing to your attention an innovative transistor architecture representing an alternative approach to next generation ultra-low power logic and known as Vertical Slit Fieled Effect Transistor or VeSFET. Here is a comprehensive document explaining the concept of VeSFET as a building block of logic circuitry.

Posted by Jerzy Ruzyllo at 01:20 PM | Semiconductors | Link

Sunday, May 4, 2014

#287 Conclusion from last five blogs

When I did comment some six weeks ago in the post #281 on the special role hydrogen is playing  in interactions with silicon, I didn't plan on the follow up five posts which discuss interactions of silicon with other elements. Well, it did happen somehow and I am glad it did because this little series of blogs sheds some light  on complex and multifaceted  interactions of silicon with the components of process ambient.


All of this is just a proverbial tip of the iceberg as all that concerns silicon with regard to its interactions with ambient is only partially or not at all applicable to other semiconductor used in the manufacture of commercial devices. Let's consider GaN, a key semiconductor in the fabrication of LEDs for lighting applications, for instance. While the effect of hydrogen in deactivation of p-type dopants (Mg in this case) is similar to silicon,  interactions of GaN with oxygen are very different in nature than those with Si. This is because with silicon oxygen reacts forming native oxide SiO2, in GaN case oxygen acts as n-type dopant substitutionally located in GaN lattice.


So, what is the message here? The message is that because of the multiplicity of semiconductor materials used these days to fabricate commercial device, each needs to be consider individually in terms of its interaction with an ambient. Otherwise, our ability to control processes would be severely compromised.

Posted by Jerzy Ruzyllo at 08:42 PM | Semiconductors | Link

‹‹ ›› is the personal blog of Jerzy Ruzyllo. With over 35 years of experience in academic research and teaching in the area of semiconductor engineering (currently holding position of a Distinguished Professor of Electrical Engineering and Professor of Materials Science and Engineering at Penn State University), he has a unique perspective on the developments in this progress driving technical domain and enjoys blogging about it.

With over 2000 terms defined and explained, Semiconductor Glossary is the most complete reference in the field of semiconductors on the market today.

‹‹ January 2019 ››
W Mo Tu We Th Fr Sa Su
1 1 2 3 4 5 6
2 7 8 9 10 11 12 13
3 14 15 16 17 18 19 20
4 21 22 23 24 25 26 27
5 28 29 30 31      

Copyright © 2019 J. Ruzyllo. All rights reserved.