Talk:Gallium antimonide

Gallium : oxidation states listed include +3 but not +2. Antimony : oxidation states listed include -3 but not -2. jimswen 09:41, 8 March 2007 (UTC)

Presently, Gallium(III)_antimonide is redirected to Gallium(II)_antimonide. If anything, should be other way around.jimswen 09:15, 8 March 2007 (UTC) GaSb is just like GaN, GaP, & GaAs stoichiometrically if not crystallographically; why change oxidation states? Sb is in column V with N,P,As, all presumably -3, because the Lewis octet is 8, and 8-5=3. Gallium is pretty strongly +3. Neither gives the other reason to deviate. The real polarity of the bonds is rather low, but oxidation number is a formalism, a mental stoichiometric accounting which need not change until the real polar content passes zero and reverses, or until one finds analogous compounds with anchoring reasons for a different oxidation-number scheme. I know of no analogous compounds other than GaAs, GaP, InAs, InSb, AlSb, etc. On the other hand, dis-analogous compounds GaTe, Ga3Sb2, and GaO would rightly be Ga(II). GaBi would probably be a metal alloy with oxidation numbers all zero. jimswen 09:31, 8 March 2007 (UTC)

GaSb is a III-V (3-5) semiconductor, Ga+3 & Sb-3, so the article should be re-named Gallium(III) antimonide. Needs to, before the table can be changed. I think it may have been a mere typo in the first place. How does one change a page name?... (jimswen 08:07, 8 March 2007 (UTC))


 * Are there any other antimonides of gallium? If not, the (III) would be unnecessary.  Chris cheese whine 21:09, 10 March 2007 (UTC)
 * Googling shows "Gallium antimonide" to be much more common than "Gallium(III) antimonide". Would that be a sufficiently precise name? -GTBacchus(talk) 05:58, 15 March 2007 (UTC)

This article has been renamed from Gallium(II) antimonide to Gallium antimonide as the result of a move request.. I know that's not what was originally proposed, but it wasn't opposed, follows WP:NC(CN), and is used by the external link. --Stemonitis

Direct bandgap value
I noticed that the cited value of the bandgap given in the lead, 0.67 eV, seems a bit lower than what I was accustomed to seeing for a room temperature direct bandgap of GaSb in the literature.

There are a few good, comprehensive reviews of the properties of GaSb. What I've gleaned from all of these secondary references is that the room temperature direct bandgap of GaSb is around 0.72 - 0.73 eV. There are a few outliers that are included in those references (ranging from 0.70 - 0.75 eV), but the recommended values they give for the room temperature bandgap seem to fall squarely in the former interval.

I'm certainly not an expert on their experimental methodology, and I don't claim to be one, but I did read into their paper a bit, and I have an inkling of an idea of what they're trying to do (they do cite a textbook in their discussion of their methods, which I have yet to track down a copy of). They apparently extracted the bandgap from absorption data by "extrapolating" a line from their absorption coefficients (as a function of photon energy), yet they do not show this line in their figure (figure 2 in the paper), which leaves me a bit skeptical as to the accuracy of their method. They also cut off their absorption measurements at about 0.7 eV of photon energy; I would want to see their absorption data at slightly higher photon energies, especially given the extrapolation procedure they claim to be doing. They're also using p-type-doped GaSb (not undoped), which may be influencing their absorption measurements as well. Depending on the doping density and the fraction of the occupied acceptor states, they may well be measuring the distance from the Zn acceptor states to the conduction band minimum, rather than the distance between the conduction band minimum and valence band maximum. If we take the Zn acceptor states as being ~0.04 eV above the valence band and the actual bandgap as 0.73 eV, one could naively expect a transition between the Zn acceptor states and conduction band minimum that is 0.69 eV, which is closer to their value reported of 0.67 eV. (I should disclaim all of this with the fact that I'm dumb, don't know anything and haven't gotten my crappy optical measurements to work yet for my own PhD project.)

Nonetheless, the value they report for the room temperature bandgap of GaSb (0.67 eV) is much lower than just about every other secondary source I've come across, so I am inclined to lean to the other, perhaps more reliable secondary sources that I discussed above, which give Eg ≈ 0.73 eV as the approximate room temperature bandgap. I have boldly updated the article to reflect this.

Sidenote: if we take the Varshni parameters for the temperature dependence of GaSb's bandgap from Vurgaftmen et. al (Eg(0) = 0.812 eV, alpha = 4.17E-04 eV/K, beta = 140 K) and assume they're valid at higher temperatures (which may or may not be the case), a temperature of T = 447 K (174 °C) gives Eg = 0.67 eV. In this sense, the bandgap of GaSb can be 0.67 eV, just at a higher temperature.

MaterialsPsych (talk) 03:06, 28 December 2023 (UTC)
 * Sounds like you've spent some time into thinking about the issue, and some sources to back up that study I cited was probably an outlier. I wouldn't worry too much about your credentials on Wikipedia - most of my experience measuring bandgaps is also crappy optical measurements. I mostly just copied the first value I found from a peer-reviewed source so the article had some content. Thanks for adding a more representative value. &#12296; Forbes72 &#124; Talk &#12297; 22:08, 28 December 2023 (UTC)