Semiconductor etc... BLOG http://www.semiconductorglossary.com/ Semiconductor etc... BLOG en <![CDATA[Tiny dots, huge potential]]> Apparently, among semiconductor apprentices there are quite a few who do not exactly get hold of what nanocrystalline quantum dots are all about. So, very briefly an attempt to underscore the key point (at least the way I see it) with regard to extreme miniaturization, or in other words, our ability to confine semiconductors in a controlled fashion to zero dimensional (typically in the range of 5-20 atoms across) crystalline particles known as quantum dots.

The key point is that the energy gap of the nanosized semiconductor dot depends not only on its chemical composition, but also on the size of the dot – smaller the dot, wider the bandgap. This is in contrast to the same semiconductor which is not 3-dimensionally confined, and hence, in which electrons can roam freely without interacting with each other. In this last case the bandgap modification can only be accomplished by altering chemical composition of semiconductor. Otherwise it remains the same regardless of whether crystalline semiconductor is in the shape of 5 mm chunk, or 300 mm in diameter wafer.
It goes without saying that the identified above unique characteristic of nanocrystalline quantum dots opens up new possibilities in devising innovative photonic and electronic semiconductor devices.
]]> Wed, 23 Sep 2009 20:40:00 +0100 <![CDATA[PV vs. semiconductors - one more comment...]]> Not long ago (May 15 and May 27, 2009) I have posted blogs in which I was arguing against the notion that the PV industry represents technical and business endeavor that is different from what‘s being referred to as "semiconductor industry" or "IC industry". I feel better about my reasoning after I came across the publication entitled "Crystalline Si Solar Cells and Micrelectronics Experience" in which IMEC‘s PV experts are proposing a roadmap for thin c-Si solar cells incorporating IC-like technologies and methodologies.

]]> Mon, 7 Sep 2009 21:34:00 +0100 <![CDATA[Is it really “post-silicon” and “post-CMOS” down the road? ]]> In my July 20th blog “Here we go again” I was commenting on silicon CMOS technology’s resilience in working its way around all sorts of perceived barriers over and over again. I also presented an optimistic view of silicon based CMOS technology, modified and altered but nonetheless based on silicon CMOS, being able to carry the load may be even all the way down to 10 nm node.

Obviously, not everybody shares my optimism. Before we agree to disagree, however, few things regarding terminology should be clarified. For instance, in one of the recent presentations no chance has been given to silicon CMOS in below-22 nm technology generations. Solutions such as gates built around nanowires are quoted for instance. But is it really a “post-Si, post-CMOS” solution? Aren’t the nanowires made out of silicon leading the charge? And configured in the CMOS structure? Or heterogeneous devices with Ge or InGaAs used to form a channel. But won’t the substrate be a 450 mm Si wafer? And a switch will be a CMOS cell? Can we really call it a “post-silicon” and “post-CMOS” solution? And graphene channels will be integrated with MOSFET (read CMOS) infrastructure formed on what substrates? Well, you’ve got the point… The way I see it the future for silicon CMOS, in one form or another, is not bleak at all, quite likely all the way to 10 nm technology. So, “people entrenched in the silicon world” don’t have to loose much sleep. At least, not yet.

]]> Wed, 2 Sep 2009 20:43:00 +0100 <![CDATA[Few questions regarding large-scale PV plants ]]> It will take a while for photovoltaics (PV) to stabilize and secure its position on the energy supply arena. Conceptually, PV is a mature technical domain. In terms of a truly large scale implementation of the solar energy conversion into electricity using semiconductor solar cells, however, what is the most efficient long-term solution is not exactly clear to me.


One way to go about it is to employ a distributed system which basically means installation of solar panels on your rooftop. Electricity generated can be used to supplement the energy intake of your household. Such a “point-of-use energy generation” makes a lot of sense to me. Alternatively, the energy is being generated by the large-scale PV plants covering very large areas with solar panels. Energy generated by such a plant goes into the grid controlled by the utility companies and then into your home.
It is obvious that both approaches are needed as the former alone won’t be able to meet energy needs of an average household. Is the latter a long-term solution, however? And a long-term I mean 20-30 years from now? Won’t some of the land covered by solar panels be better used to grow crops? Won’t the price of the land start factoring in eventually? Won’t there be an adverse impact on the environment? Thousands of square miles of land essentially shielded from the sunlight? And how about the environmentally friendly, energy efficient disposal of the tones of solar panels once their performance will deteriorate beyond any reasonable limits? Particularly, if solar cells are made out of not very "friendly" materials (look beyond silicon here…)? I am sure PV community has answers to those questions. Personally, I am looking at this time for the efficient rooftop PV panels rather than counting on my utility company to send my way electricity generated by a large-scale PV plants.
]]> Sun, 23 Aug 2009 21:14:00 +0100 <![CDATA[Let’s be cautious deciding on what is “hot”… ]]> Many years in an advanced semiconductor research during which I kept an eager eye on the “hottest” developments in semiconductor science and engineering taught me a few lessons. One of them is that we should be very cautious deciding on what’s really "hot" among the new, potentially groundbreaking developments before they prove themselves in real-life applications.

I recall quite a few promising breakthrough solutions which down the road displayed unexpected fatal shortcomings. A classic example is a case of high-temperature superconductors which we all thought would resolve problems of interconnect lines in cutting edge integrated circuits for many technology generations to come. Well, it turned out that materials in questions are losing their superconducting properties at the current densities needed to run advanced ICs. The result: no go… In the similar fashion diamond did not live up to our very high expectations (see my blog of Jan. 19. 2008 “Diamond semiconductor not so shiny”).

I am sure other examples of similar situations can be produced. At this time, however, I suggest we keep our fingers crossed so that promise light-emitting diodes hold in lighting applications won‘t disappear because of the nagging inherent problem – at the current levels needed for general lighting the LEDs fail to deliver (see cover story in IEEE Spectrum, No.8, 2009).
All this should be kept in mind at the time when so much, and in so many diversified areas, is going on in semiconductor science and engineering and expectations regarding some technical solutions are so high..
]]> Tue, 18 Aug 2009 20:03:00 +0100 <![CDATA[Engineering vs. PR]]> I don‘t recall any segment of semiconductor industry getting such a high dose of PR attention as photovoltaics (PV) are getting these day (yes, PV sector is a part of semiconductor industry - see my blog of May 15, 2009). All this is clearly directed toward business community and is probably meant to encourage new investments in the PV industry.

And it’s all fine except that too much of the aggressive PR may actually be counterproductive. What if it starts looking like an act of desperation? There is no doubt PVs are bound to grow. Ultimately, however, the extent of the growth will be determined primarily by “$$/watt” factor, or in other words by the cost of energy extracted from the sunlight through photovoltaic effect in semiconductors. And no PR will be able to help it on the long run if the cost of energy produced this way won’t be competitive. Remember solar cell mini-frenzy during the energy crisis in the ‘70s?

Besides, quite often products of the PV’s PR efforts are annoyingly full of air and short on substance. Not to mention frivolous negligence of technical accuracy displayed in some publications (e.g. reference to thin-film and crystalline silicone solar panels in the Special Advertising Section, Aug.3, 2009 issue of Newsweek).

In conclusion, what if some of the PR funds would be diverted to increase support of the engineering efforts in photovoltaics? Just a thought… .

]]> Thu, 6 Aug 2009 18:56:00 +0100 <![CDATA[CCD – yet another key technology based on semiconductors]]> “CCD” is an acronym commonly used in the reference to video and still cameras (it stands for Charge Coupled Device). There seem to be a recognition of its function as a vital image sensing part of the camera, but what exactly “CCD” is, does not seem to be clear even to those who freely brag about advantages of 10 megapixel camera over 8 megapixel one (just ask any salesperson in any electronics store). On the internet it is easy to find both simple definition of a CCD or a broader description of its operation.

Here, in the spirit of the mission to present semiconductors as the materials playing pivotal role in the growth of our technical civilization, let me just point out that the CCD technology is based entirely on the pure breed semiconductor material, silicon to be exact.

By the way, an alternative to CCD image sensors is a CMOS based technology. Another silicon based workhorse…...
]]> Tue, 28 Jul 2009 20:12:00 +0100 <![CDATA[Here we go again]]> My adventure with semiconductors spreads over the larger number of IC technology generations that I would care to admit. It all started about 10 µm design rule and in the not too remote future is bound to complete a three orders of magnitude reduction of transistor geometry. In the process I witnessed several cycles during which “insurmountable physical barriers”, “red-brick walls”, “limits of technology”, “ends of the road”, etc., etc., were worked around, broken, smashed, and otherwise left in the dust by the relentlessly advancing silicon IC technology.

In the case you didn’t notice we are now in the midst of the similar cycle. Not long ago any mass produced transistor with a gate length shorter than 100 nm was considered as something close to a technical impossibility. Am I mistaken or as of today a 32 nm technology is essentially a done deal, and 22 nm is coming? Beyond that 16 nm is considered to be feasible, and 10 nm is not seen as an unreachable goal? And all this based on the good, old silicon, modified and supported by other materials, but without major revolutionary changes in the process technology? Well, all we have to do is to wait and see...

]]> Mon, 20 Jul 2009 10:19:00 +0100 <![CDATA[Summer travel]]> This is probably the longest "no-new-blog" strech since I started blogging some 1.5 year ago. This is all because of the Summer travel. But I‘ll be back....

]]> Tue, 7 Jul 2009 04:26:00 +0100 <![CDATA[Displays go semiconductors]]> Out of our five senses the vision is the most useful when it comes to the exchange of information with a digital domain. Hence, displays are an indispensible link between humans, computers and other numbers crunching devices. This status quo will remain in place for quite some time. (That is until we will be able to couple digitally processed information directly to our brains bypassing our five sensing systems). The question is how do the semiconductors fit into a growing display business?

Not long ago display technology was dominated entirely by bulky and power hungry CRT (Cathode Ray Tube) imaging devices. The working of a CRT has nothing to do with semiconductors. In the currently dominating active matrix liquid crystal displays (LCD), however, semiconductors are playing pivotal role in the form of thin-film transistors (TFT) controlling each pixel. The next step is a generation of displays based entirely on semiconductors. This is because not only transistors, but also light emitting elements of the display are semiconductor based Light Emitting Diodes, LEDs. (Fabricated using either organic -or inorganic semiconductors). So, slowly but surely, displays are becoming yet another prominent member of the semiconductor devices family. And the role of semiconductors in our lives is growing and growing…..

]]> Mon, 15 Jun 2009 07:20:00 +0100