About Silicone and Siloxanes

It has been an established fact, ever since these substances were invented, that “Silicone” did not mean the same thing as “Silicon”.

The word without the ‘e’ at the end, stands for an element in the periodic table, and is therefore a type of atom.

‘Silicone’ always referred to a polymer, which typically contains the elements Silicon, Oxygen, Carbon and Hydrogen, although certain practical preparations of Silicone, may contain additional elements such as Sulfur… This polymer forms chains, in which Silicon atoms alternate with Oxygen (bridging) atoms. Therefore, the Silicone -family of polymers is also referred to as ‘Siloxanes’, which sounds substantially different from ‘Silicon’.

Typical Silicone ‘Cocking Compound’ would be a practical example of the polymer.

I think that the fate which this terminology eventually succumbed to was twofold:

  1. It was always impossible to explain what the difference was supposed to be, to laypeople, and
  2. The spelling of the two words was so similar, that laypeople who saw no reason to distinguish between them in the first place, would also use these two words interchangeably.

It might be, that the modernistic way to refer to the polymer, is just to call it ‘Polysilicon’. (?)

Even though the reader might have hoped for more complexity, there are really only two pathways by which Siloxanes polymerize. They respond to moisture, and either give off Hydrogen Chloride gas, or Acetic Acid, the latter of which also gives Cocking Compound its vinegar-like odor. This would happen in both cases, either because a Chlorine atom, or an Acetate group was attached to the Silicon atoms. With water, the corresponding acid develops, and the Oxygen atom of the water forms the bridge, between two Silicon atoms.

Dirk

 

Print Friendly, PDF & Email

Major System Update Today, Downtime

I go the unusual route, Of hosting my Web-site and therefore also this blog, on one of my own private computers at home, that is named ‘Phoenix’. Therefore, the availability of my blog is only as good, as that of my PC.

This evening the Debian / Jessie repositories pushed through a major update, which turned my Debian 8.10 system into a Debian 8.11 system. Additionally, this included an update to the proprietary, legacy nVidia graphics driver, version 304.x , which is pretty much the only device-driver left, which will work with an old on-board graphics chip, which was the:

GeForce 6150SE nForce 430/integrated/SSE2

Hence, if after the update, the X-server had not restarted, I would pretty much have been finished with this computer!

Luckily, the graphics-driver update seemed to take well, and the whole procedure seemed to go smoothly.

Due to the reboot which was required, my site and blog were off-line, from about 20h05 until about 20h10. Because this was only for 5 minutes, I’m sure this did not inconvenience my readers much.

Dirk

 

Print Friendly, PDF & Email

Intrinsic Silicon

Many people already understand, that two types of silicon exist, N+ -Doped, and P -Doped.

Well I’ve known for some time, that another type of silicon which exists, is called ‘Intrinsic Silicon’. This is a form of silicon, which theoretically contains no dope at all, and which is therefore non-conductive. It’s not even a semiconductor in that state.

This type of silicon might be of some interest in the design of modern Integrated Circuits, especially in the reduction of the capacitance of individual transistors. But there areĀ  essentially two problems with its use:

  1. It’s practically impossible for the silicon to be perfectly pure. The concentrations of Dope, in the N+ or P -Doped silicon, are already extremely low. The concentration of impurities in Intrinsic Silicon is simply lower, industrially, than in the intentionally-doped silicon, not truly zero. And what this means in practice, is that ‘larger pieces’ of Intrinsic Silicon are still partially conductive. In fact, how low the concentrations of N+ or P -Dope can be brought in the industrial process, depends on how low the level of impurities is, in the silicon, to begin with. In either type of intentionally-doped silicon, the concentration of dope must still be at least one order of magnitude greater, than the level of impurities was.
  2. Actually, I think that Intrinsic Silicon is more expensive in bulk, than either type of intentionally-doped silicon, which means, that if the entire wafer needed to be made out of it, since the substrate of the wafer is meant to provide mechanical support as well, then the cost of the manufacturing process would increase.

Yet, small pieces of Intrinsic Silicon, as the following image shows, can still be used to provide lateral insulation, between the P -Doped and the N+ -Doped wells of individual transistors, where a “buried oxide layer” provides vertical insulation between those wells, and the actual wafer:

https://upload.wikimedia.org/wikipedia/commons/e/ee/Cmos-chip_structure_in_2000s_%28en%29.svg

And, it would be my expectation that because Intrinsic Silicon is ‘non-conductive’, larger pieces of it should also be optically transparent, which means that some people might mistake it for glass.

By definition, glass would be ‘amorphous’, which means ‘not crystalline’, which would make actual glass useless as a semiconductor. However, amorphous forms of silicon can readily be used in the design of wafers, as long as they do not need to participate in the actual semiconductive behavior between N+ and P -Doped silicon.

Dirk

 

Print Friendly, PDF & Email

A Basic Fallacy which Gets Taught in Circuit Design

There is a type of MOSFET circuit, which Electronics Students are taught to use, which gets mislabeled a ‘Current Mirror’, and the way it gets diagrammed, is like so:

currentmirror_1_c

An implicit promise which this makes is, ‘R1 will control how much current flows through M1, because the supply voltage is high, and the same amount of current will always flow from the Drain of M2, to the Output, no matter what the Output voltage is. Because why else, would this get called a current mirror?’

And an interesting design-feature which we may observe, is that by defining the Gate-Substrate voltage of ‘M1′, the current which is drawn through ‘R1′, also controls the Gate-Substrate voltage of ‘M2′, which may lead some people to conclude that indeed, the Source-Drain current of ‘M2′ will be equal to that of ‘M1′ exactly, all the time.

The main reason for which I see this used, is the fact that Its equivalent worked well, with Bipolar Junction Transistors, where Collector current was in fact controlled well, by Base-Emitter voltage. But a MOSFET has some crucial differences, to a BJP. One such difference is the fact that By itself, the Gate-Source voltage will not be enough to define the Source-Drain current ! ( :1 )

I was recently reminded myself, that both voltages: Drain- and Gate- , together define Drain current. ( :2 ) In fact, what I would guarantee, is that If the Substrate voltage was in fact disconnected from the Source, the MOSFET will act as a variable resistor, the value of which, between its Source and Drain, will be defined by the Gate-Substrate voltage. Only, If the Substrate voltage is in fact tied to the Source – as it often is – Then, it could only be used as a variable resistor, for very small signals, because larger signals do also appear at the Source, and therefore also affect Gate-Source voltage, which would cause distortion when used in such a way.

In fact, the only guarantee which the above circuit makes is, ‘If the Threshold Voltage of the MOSFET (‘VTO’) is unknown, then larger or smaller currents flowing through R1, will also lead to larger or smaller currents, but always non-zero, to the Output. Very probably however, the current flowing to the Output will be greater, than the current through R1 was, since the voltage of Output, is unlikely to approach (Vcc – VTO).’

(Updated 06/24/2018, 22h20 : )

Continue reading A Basic Fallacy which Gets Taught in Circuit Design

Print Friendly, PDF & Email