Talk:Schottky diode

REVERSE BIAS USE is quite common
The author was reading about power use and transistor types. The entire article was neglecting to talk about the use of the low and stable and specifiable reverse bias break down voltage as a voltage limiter. You can buy a 20 volt Shottky, which means that its reverse bias break down voltage is 20 volts, and it can be used to bypass any higher voltage THROUGH the diode. There can be stablity issues, but as noted the Shottky is quite fast and its stable up to eg PWM, frequencies.

220.233.121.43 (talk) 00:23, 8 November 2015 (UTC)

This (and corresponding part of the main article) seems wrong! Someone may have confused a Schottkey diode with a Zener diode.

86.26.2.112 (talk) 16:09, 19 November 2015 (UTC)


 * It would perhaps therefore be helpful for someone, with considerably greater expertise than myself, to include a few sentences/a paragraph or two, highlighting the differences (and similarities) between the Schottkey (?sp) and Zener diodes for purposes of disambiguation and to rationalise better, and in more detail, potential understanding.
 * Furthermore, is ‘ohmic contact’ distinct from a ‘de facto resistor? And, if so, in what way? [this should provide evidence for my aforementioned lack of expertise] Tweet7 (talk) 10:06, 18 April 2023 (UTC)
 * If you evaporate metal onto a semiconductor, it either forms a Schottky barrier diode or an ohmic contact. Mostly, metals that diffuse into the crystal, and so make a not so sharp transition, make ohmic contacts. That is meant to be in contrast to a rectifier, which is definitely not ohmic. This is similar to the way that a fuzzy transition in index of refraction reduces reflections. Gah4 (talk) 07:04, 19 April 2023 (UTC)
 * If you evaporate metal onto a semiconductor, it either forms a Schottky barrier diode or an ohmic contact. Mostly, metals that diffuse into the crystal, and so make a not so sharp transition, make ohmic contacts. That is meant to be in contrast to a rectifier, which is definitely not ohmic. This is similar to the way that a fuzzy transition in index of refraction reduces reflections. Gah4 (talk) 07:04, 19 April 2023 (UTC)

Sentences in conflict
'typically 300mv' conflicts with 'Schottky diodes have a drop of only about 0.2 volts.' in opening paragraph.
 * Fixed.--Heron 19:26, 21 Jun 2005 (UTC)

This article needs expanding with more detail and facts on metal-semiconductor contacts. Any takers?? Al


 * As well as I know, they are between 0.2V and 0.4V, depending on the specific part number and current rating. Gah4 (talk) 18:15, 30 March 2020 (UTC)

Schottky Resistance
Whilst the diode does require lowered voltage this in turn means a reduction is temperature and so an increase in later resistance

Spelling
Is it Schottky or Shottky? There is the Walter Shottky institute in Germany. Can anyone verify his real spelling? —The preceding unsigned comment was added by 65.209.67.191 (talk) 20:51, 13 February 2007 (UTC).


 * S. M. Sze (Physics of semiconductor devices, ISBN 0471056618 ) mentions the "Schottky effect", and B. Jayant Baliga (Power Semiconductor Decices ISBN 0534940986 ) has a section on "Schottky Barrier Rectifiers". CyrilB 21:32, 13 February 2007 (UTC)

Pronunciation
-how do you pronounce 'schottky'?

- SHOT-key - RC Cola 20:35, 29 May 2006 (UTC)
 * ch !=> h; k !=> k. It's Skotecy. -lysdexia 02:14, 4 October 2006 (UTC)
 * IMHO you're wrong; I speak German and SHOT-key seems to be right. sch [ger] = ʃ [ IPA ] = sh [eng]; k is pronounced similarly in both languages. Torzsmokus 16:39, 27 January 2007 (UTC)

Silicon Carbide devices by Cree
Cree Inc. makes SiC schottky diodes with high Vr ratings. Specs for CSD20120 are: Vr 1200V, If(avg) 20A, Vf(typ) 1.6V @ 10A. CSD20120 Datasheet (PDF). &mdash;Ryan 23:45, 29 June 2006 (UTC)


 * Yes, it should be introduced alongside of the limitations of Si schottkies (low temp and low voltage). CyrilB 07:49, 30 June 2006 (UTC)

Schottky diode speed
It should be corrected that microwave mixers are commonly constructed using silicon schottky diodes for operation past 65 GHz. GaAs schottky diode mixers can be used well into the millimeter wave part of the spectrum...the wiki article indicates only to 5 GHz.


 * There are a few applications "just a little" beyond either 5 or even 65Ghz
 * http://www.simics.tec.ufl.edu/papers/sankaran3.pdf -- a schottky with an estimated cutoff of 400Ghz
 * http://trs-new.jpl.nasa.gov/dspace/handle/2014/8909 -- a schottky diode used within a device at the 1400-1900 Ghz range
 * http://cat.inist.fr/?aModele=afficheN&cpsidt=3433020 -- a reciever operating in the 250Ghz range
 * Andy t roo (talk) 14:13, 24 September 2008 (UTC)

Applications
Why are there two applications sections? Nano Dan (talk) 20:54, 12 July 2008 (UTC)
 * I combined them into one, but the content needs to be re-written to be more coherent and with less repeating.. probably expanded too... Nano Dan (talk) 00:37, 17 July 2008 (UTC)

Graph
What about a V/I graph? I find those very useful. 151.61.26.186 (talk) 16:21, 17 March 2009 (UTC)

Wrong physics
The article now says "When current flows through a diode, it has some internal resistance to that current flow, which causes a small voltage drop across the diode terminals. A normal diode has between 0.7-1.7 volt drops...." Resistance doesn't correspond to a voltage drop without a specified current, so this must be wrong. I'm removing the wrong part, but something informative should be inserted 71.141.88.244 (talk) 23:52, 23 April 2009 (UTC)


 * I think this sentence is using the word "resistance" in the colloquial sense. I like the edit though. --Steve (talk) 23:56, 23 April 2009 (UTC)

Yes. Well in formal text, it might be possible to refer to dV/dI. the ratio of the change in voltage for the change in current. This might confuse readers into thinking that higher level mathematics has been applied and there  is a function that has been d/dx'd.   So one might call it a slop, however with no voltage current curve, the author didn't want to refer to slope.. But yes you can call it resistance if   Voltage =  Diode drop + IR....  Or "impedance"  is often used to refer to non-Ohmic  operation... back EMF ? Hehe same thing. — Preceding unsigned comment added by 220.233.121.43 (talk) 00:20, 8 November 2015 (UTC)

Disaster zone
Too many editors who are not competant or experts. Needs a major clean up and re-org. It is such a fundemental component, deserves better. I really really do not like mention of commercial companies in the body of the article. A reference should be made to a published scientific article, not a commercial brochure or a copany web site. I will come back when have a chance. —Preceding unsigned comment added by 12.229.112.98 (talk) 22:10, 19 February 2010 (UTC)

Agreed. Schottky barriers are damn cool things. If you read this article and the Schottky barrier article, you get the impression that the M/S junction just has some kind of rectifying feature with low voltage drop. This completely ignores the fact that way the Schottky barrier works, and the source of it's useful features, is quantum tunneling; hence the lower voltage drop (tunneled particles maintain kinetic energy; what voltage drop does occur is due to the reflection of some the particles at the potential barrier, and the redistribution of what charges make it through the M/S junction ). Although the Schottky barrier height is poorly understood, The Schottky Diode is an absolutely excellent example of how quantum physics effects the macro world, and how we can use the stranger features of quantum mechanics in many pieces of modern technology.

This article needs a rewrite by a physicist, not an electrical engineer. —Preceding unsigned comment added by 128.138.96.92 (talk) 21:35, 24 February 2011 (UTC)

Request for Discussion of in electronic articles
Please see the corresponding discussion thread at Wikipedia talk:WikiProject Electronics. Thanks! • Sbmeirow  •  Talk  • 23:34, 15 December 2013 (UTC)

Comments on the article
1. "This can be used for simultaneous formation of ohmic contacts and diodes, as diodes form between the silicide and lightly doped n-type region and ohmic contacts form between the silicide and a heavily doped n- or p-type region."

Too many ands ... the sentence should be improved.

2. "They also have forward voltage closer to that and require more attention in layout."

What is "that"? The pn junction voltage?

3. "The junction lies in direct contact to the thermally sensitive metallization, a Schottky diode can therefore dissipate less power than an equivalent-size p-n counterpart".

The first part of the sentence does not connect well with the rest of it. It needs to be improved.

4. "This "instantaneous" switching is not always the case."

I think there is a better way to phrase this.

5. "The Schottky diode models distributed by manufacturers are also often given without a valid TR parameter leading to an optimistic expectation."

Is that supposed to be trr?

6. "(BV, IR)".

What is BV? What is IR?

7. Germanium diodes are mentioned in the article. Then they are labeled as superseded. Are they actually used at this time? Isn't germanium technology obsolete? ICE77 (talk) 03:20, 25 July 2015 (UTC)
 * I don't know specifically, but for a long time Ge diodes with their natural 0.4V voltage drop were used in some applications, especially low current applications. Si Schottky diodes at usual current levels are about 0.4V, so a Ge Schottky diode might be even less. There are also SiGe alloys, which might be used for some devices. Gah4 (talk) 18:19, 30 March 2020 (UTC)
 * I don't know specifically, but for a long time Ge diodes with their natural 0.4V voltage drop were used in some applications, especially low current applications. Si Schottky diodes at usual current levels are about 0.4V, so a Ge Schottky diode might be even less. There are also SiGe alloys, which might be used for some devices. Gah4 (talk) 18:19, 30 March 2020 (UTC)

Beginning para reversions
Once again, User:VQuakr, you are reverting my edits for no good reason. You reverted the bold lines. Please explain why those lines do not belong in the article.

Before I began work on the article, the paragraph originally read:

The above paragraph contains material that is complete rot. The current does not cause the voltage to appear, the breakdown voltage is applied and a nano second later the diode becomes conductive. To say otherwise is to put the cart before the horse. Yet you reverted my edit, suggesting that you agree with the original writing. I wonder, do you know anything of how diodes work? Zedshort (talk) 16:06, 10 May 2016 (UTC)
 * , I think if you added that information back with a reference supporting it, it would be unquestionably fine. Toddst1 (talk) 14:13, 11 May 2016 (UTC)
 * it's not really a sourcing issue, more editorial. Of the two proposed sentences, the first is true for all diodes and unnecessary in the lede for an article on this type of diode. Similarly, the 2nd proposed sentence regarding reverse breakdown voltage is true for all diodes. If you think additional general discussion in the lede is warranted, let's discuss how to improve the phrasing into something less conversational before it is re-added. VQuakr (talk) 18:49, 11 May 2016 (UTC)
 * I'm puzzled by your definition of editorial content. Adding a line of explanation of how such devices works in the form of an analogy is hardly editorializing. A statement that is true for all diodes is true for this diode and hence is of use here. Adding a single line, does not constitute a "general discussion". If there was a very large body of writing that cluttered the intro I could see your objection. But really, a line? Please stop running interference on my attempt to make these articles, readable, understandable, and relatively self-contained. I suspect that this controversy is more about your pride than an attempt to improve things. You don't like my edits on the battery article and so you follow me to other articles to revert what I have done. Your behavior comes across as vengeful. I suggest you disengage your viscera and re-engage brain. Zedshort (talk) 19:47, 11 May 2016 (UTC)


 * I too am puzzled by 's objection. The wording seems fine to me.   If the statements are so obvious that it doesn't need sourcing, I see no reason not to have it included/restored.  Toddst1 (talk) 22:03, 11 May 2016 (UTC)


 * I see no problem with using a one-way valve analogy. It is common.  There are problems with other parts of the edit.  Specifically, there is no sufficient forward voltage to cause current to flow in the forward direction.  Any forward voltage will cause forward current.  Forward current is a smooth continuous, nearly exponential, function of forward current voltage.  The manufacturer’s forward voltage specification is simply the voltage required to cause a certain current, typically 1 mA on small devices but could be larger currents on larger devices.  Constant314 (talk) 23:15, 11 May 2016 (UTC)


 * This puzzles me beyond expression: "Specifically, there is no sufficient forward voltage to cause current to flow in the forward direction. Any forward voltage will cause forward current." I have hardly a clue what the first sentence was supposed to mean, and the second was pure rot. A diode is non conductive until it a sufficient voltage is applied. First a sufficient voltage is applied, then the conductivity drops, rather like a dam is lowered and water is allowed to spill through. There's another analogy for you to object to. If the voltage is removed/dropped below the breakdown voltage, its resistance shoots up to essentially that of a non conductor, i.e. the dam pops back up and the current stops flowing. There's another analogy for you to object to. The breakdown voltage is a function of the type of diode not the size of the device. Have fun with the article, this a waste of my time. Zedshort (talk) 03:11, 13 May 2016 (UTC)


 * All diodes leak some current at any nonzero potential, including Schottky diodes below their knee voltage. It's just a relatively small amount of current. I suspect intended a meaning similar to, "Specifically, there is no minimum forward voltage necessary to cause current to flow in the forward direction." The water analogy works great for teaching Ohm's law but breaks down for A/C and semiconductors, with diodes being a marginal exception. VQuakr (talk) 05:57, 13 May 2016 (UTC)


 * The water analogy can be a useful teaching tool; it's just out of place in the second paragraph of the lede in an article on a particular type of diode. Sort of like if I added "a camel is rather like a lumpy horse" midway through the lede of Camelus bactrianus. We write in summary style; diode should contain the high-level info. This article's lede should focus on the actual topic. Beyond that, though, by juxtaposing statements about diodes in general with information unique to the topic we risk confusing the reader. If we think a paragraph about diodes in general is really necessary for the lede, let's at least make it a coherent paragraph. VQuakr (talk) 01:39, 12 May 2016 (UTC)
 * I see your point. A succinct lede about a particular type of diode doesn’t need a general discussion about diodes.  Any reader looking up Schottky diode probably already knows what a diode is and if he doesn’t then there is the wiki-link to the diode article.Constant314 (talk) 16:37, 12 May 2016 (UTC)


 * You have quite the ability to obfuscate and cast derision on other's work. Let's see you actually do some decent writing. Zedshort (talk) 03:11, 13 May 2016 (UTC)

Applications
I've added sample and hold circuits as another application. (with reference) TheUnnamedNewbie (talk) 09:19, 9 January 2017 (UTC)


 * Thanks for the addition. Did the source mention any disadvantages? Constant314 (talk) 15:04, 9 January 2017 (UTC)


 * I'm affraid not, but I'll look into it in more detail once I'm studying the topic again next week. TheUnnamedNewbie (talk) 09:03, 11 January 2017 (UTC)

polyacetylene
Someone seems to keep wanting to add mention of polyacetylene diodes. Certainly that is possible, there has been work on organic semiconductors for many years, and organic LEDs seem also to exist, but I don't believe it is the usual Schottky diode. Even so, if such devices exist, they might have a section in the article. Gah4 (talk) 18:23, 30 March 2020 (UTC)

Reverse recovery times: Schottky vs pn diodes
Small imprecision in the section Reverse recovery time. The text states in Schottky diodes there is nothing to recover from (i.e., there is no charge carrier depletion region at the junction). Actually there is charge depletion on the semiconductor side of a Schottky barrier and it depends on the bias, just like in a junction diode. Both pn and Schottky diodes have a junction capacitance C_J. The big difference is that pn diodes also have minority carrier diffusion and a diffusion capacitance C_D due to the stored minority carriers. In forward bias C_D is way larger than C_J and the true reason why pn diodes are slow: it takes a lot of time to remove those minority carriers and the diode stays ON for quite a while (on the scale of the recombination time). Possible reference: R.F.Pierret "Semiconductor Device Fundamentals", Chap.8 - pn Junction Diode: Transient Response. I am not editing the main article since I have never done that: I don't want to mess up! Wikirod76 (talk) 13:10, 19 June 2020 (UTC)
 * Looking at the data sheet for the 1N5817, reverse recovery time is not indicated. But yes, I don't believe that it is zero. There is some discussion (I forget where) on doping of SB diodes. If you dope too high, you get an ohmic contact. (That is, tunneling through the barrier.) Too low isn't good, but I forget why. Maybe it is too slow. In any case, there is a just right.  PN junction diodes have a nice continuous, low recombination site, junction. I always figured that somewhere in the metal-semiconductor junction there were good recombination sites, but never tried to get into more detail. If the diodes were really good, we would use them more.  Gah4 (talk) 20:57, 19 June 2020 (UTC)
 * Yes exactly: recovery time is not zero since there is a junction capacitance, even in Schottky diodes.... but C_J can be made small by properly designing the junction geometry, like in point-contact diodes. Reducing the additional C_D in a pn diode is less obvious, I remember a book discussing that but I can't recall the details, you probably want a relatively fast recombination time so a less than ideal (very doped?) semiconductor. Anyway the fundamental concept is clear: to speed up a pn you need to store less minority charge and/or be able to somehow remove it faster. Large doping in Schottky will reduce the width of the SB (thus the size of the *existing* depletion region) and lead to electron tunneling, I definitely agree with that. Wikirod76 (talk) 12:26, 21 June 2020 (UTC)
 * Yes exactly: recovery time is not zero since there is a junction capacitance, even in Schottky diodes.... but C_J can be made small by properly designing the junction geometry, like in point-contact diodes. Reducing the additional C_D in a pn diode is less obvious, I remember a book discussing that but I can't recall the details, you probably want a relatively fast recombination time so a less than ideal (very doped?) semiconductor. Anyway the fundamental concept is clear: to speed up a pn you need to store less minority charge and/or be able to somehow remove it faster. Large doping in Schottky will reduce the width of the SB (thus the size of the *existing* depletion region) and lead to electron tunneling, I definitely agree with that. Wikirod76 (talk) 12:26, 21 June 2020 (UTC)
 * Yes exactly: recovery time is not zero since there is a junction capacitance, even in Schottky diodes.... but C_J can be made small by properly designing the junction geometry, like in point-contact diodes. Reducing the additional C_D in a pn diode is less obvious, I remember a book discussing that but I can't recall the details, you probably want a relatively fast recombination time so a less than ideal (very doped?) semiconductor. Anyway the fundamental concept is clear: to speed up a pn you need to store less minority charge and/or be able to somehow remove it faster. Large doping in Schottky will reduce the width of the SB (thus the size of the *existing* depletion region) and lead to electron tunneling, I definitely agree with that. Wikirod76 (talk) 12:26, 21 June 2020 (UTC)

Ok I edited the main article. Stating that Schottky diodes have no charge depletion (and the consequent junction capacitance) is incorrect. The reason why pn diodes are slow is that they have a (typically large) diffusion capacitance. In other words, they accumulate plenty of minority carriers in their diffusion region during the ON state, and it takes a lot of time to remove them. I added the reference too. The section might still need some clean up... my edit is somewhat repeated (majority carrier device, diffusion etc) in the last paragraph of the section (which was in contradiction with what stated in the first paragraph before my edit). I also notice that 200kHz-2MHz is indicated as a good range for "high-speed diodes", but Schottky can keep working well into the microwave spectral range..... Wikirod76 (talk) 12:07, 21 June 2020 (UTC)
 * You can pretty much say that anything that claims to be zero is wrong, I suppose with the exception of superconductors. I seem to remember gold doping also for increasing recombination and speeding up diodes. Gah4 (talk) 22:57, 21 June 2020 (UTC)
 * You can pretty much say that anything that claims to be zero is wrong, I suppose with the exception of superconductors. I seem to remember gold doping also for increasing recombination and speeding up diodes. Gah4 (talk) 22:57, 21 June 2020 (UTC)

History section does not describe anything about the diode
The History section only describes the inventor and another of his discoveries and nothing about the development or discovery of the diode. The article on Walter Schottky also describes nothing about the history of the diode.

Can anyone expand this section?Lkingscott (talk) 10:36, 11 March 2023 (UTC)