A New Picture of Superconductivity : Lightning Bolt Electrons in a Crystal


An electron is an electron is an electron . That is , all electrons are the same fundamental particles of matter , wherever they are . However , they behave differently in different environments . In that respect electrons can be placed in either of two classes : electrons associated with atoms , and free electrons (those that aren't associated with any specific atom) .

Before proceeding any further , I should advise the reader that this essay is an attempt to describe superconductivity in chemistry , not physics terms , as I am by education and training a chemist (now in retirement a writer , no longer a practitioner) and not a physicist . For example , if a condensed matter physicist should ask me to design a new superconducting compound based on Majorana fermions , he would be met only with an uncomprehending , blank stare . On the other hand , if I was asked to formulate a new superconducting material based on valence shell electrons and orbitals (and occasionally hypervalence ones as well) , then all right , now we're in business ! And that's what this web page is all about : a chemistry approach to the synthesis of a series of new chemical compounds that I think have a good chance of being high temperature superconductors ; in fact , they may even be superconducting from Absolute Zero all the way up to their melting or decomposition points (in other words , all-temperature superconductors) .

Various types of free (unassociated) electrons , mostly of the static electricity kind , are well known . These electrons are outside matter and cling to the surfaces of molecules , probably by van der Waals dipolar bonding . We sometimes see static sparks fly off clothing and other textiles , especially when the air is very dry in Winter ; and everyone is familiar with lightning , which is always spectacular . For example , see this impressive YouTube video of a recent (October 26 , 2014) , extremely violent electrical storm over Adelaide , Australia [MP4 , 7063 KB , runtime 1:45] . Tesla coils can produce artificial lightning bolts :

The above picture was copied from the Wikipedia web page , Tesla coil . My thanks to the author of this photograph , and Wikipedia , for implied permission to reproduce it on this web page .

I used to have a small , handheld , laboratory version of the Tesla coil for verifying the effectiveness of a vacuum inside a glass apparatus . It could produce sparks several inches long . Obviously , the electrons in static electricity , lightning , corona discharges , and Tesla coil sparks are outside matter and are therefore free in nature . Three more familiar types of free electrons are those in fluorescent light tubes , cathode ray tubes (like my old CRT computer monitor on which I'm typing these words) , and the electrical arcing noticeable when a power cord attached to a powered appliance is pulled out of the wall receptacle .

The electrons inside matter are virtually all associated electrons ; that is , they are bonded tightly to their parent atomic kernels . The electrons that flow as electricity in a conventional electrical conductor are associated electrons . In common metallurgical metals the valence shell s and p orbitals can overlap continuously throughout the crystal lattice of the metal to form the s,p conduction band . If this s,p conduction band is partially filled with valence electrons they will be able to drift through the s,p orbitals under an applied potential difference (p.d.) . The material will be an electrical conductor and the s,p conduction band will be its metallic bond .

The energies of the free , mobile electrons in the metallic bond , because they are in the s,p orbitals , are subject to the Fermi-Dirac distribution (or to Fermi-Dirac statistics , if you prefer) . This is because the s,p orbitals form a sort of lattice-wide orbital over the spatial dimensions of the crystal . I have referred to such a polymerized s,p orbital as a crystal orbital (abbreviated as XO) in previous Chemexplore web pages . The crystal orbital is synonymous with the metallic bond and conduction band , and reflects the reality that in any given specimen of a metal there is only one single metallic bond in it .

However , the Pauli Exclusion Principle requires every orbital electron to have its own unique set of coordinates , both in physical and in energy space . The analogy is that every telephone in the world must have its own unique phone number , and every website on the Internet must have its own unique web address (URL) . By having unique Pauli Exclusion coordinates the orbital electrons will avoid being in the same physical and energy spaces at the same time . In the giant blob” XO the metallic bond free electrons are all in the same physical space , so they must be distinguished from each other in energy space . This differentiation is achieved by placing each of the free electrons in its own energy level (that is , they have the same space address , but different energy addresses) . That's where the Fermi-Dirac energy level distribution takes place . A good discussion of the Fermi-Dirac distribution can be found in the article by A.R. Mackintosh , “The Fermi Surface of Metals”, Scientific American 209 (1) , pp. 110-120 (July , 1963) [DOI] ; also in the excellent overview of the solid state chemistry of metals by W.J. Moore , Seven Solid States , An Introduction to the Chemistry and Physics of Solids , W.A. Benjamin , New York , 1967 [ABE] ; Ch. 2 , “Gold”, pp. 41-72 , especially pp. 47-51 .

The associated electrons in conventional electrical conduction are tightly controlled by their respective atomic kernels . Even though the free electrons can flow in the lattice within the XO , they are still firmly bound within it . What happens to release the associated electrons so they can form the Cooper pairs in the superconducting state ?

In the earlier Inert Pairs web page I considered the case of mercury , frozen in a bath of liquid helium at 4 K . At that temperature it suddenly becomes superconducting . At slightly higher temperatures it becomes a normal electrical conductor again . The electronic configuration of elemental mercury is very simple : it's 6s2, which are its outermost valence shell electrons . Mercury has a 5d10 shell of electrons , but in the IIB/12 Family (zinc , cadmium , and mercury) the nd10 electrons seem to be submerged” in energy relative to the (n+1)s orbitals , and under normal conditions never participate in any sort of chemical bonding . The IIB/12 elements have empty , but electronically active outer p orbitals that do participate in chemical bonding . They also form the s,p conduction bands in zinc , cadmium , and to a minor extent in mercury , whose inert pair effect is particularly pronounced (making it a liquid metal at room temperature and a rather poor electrical conductor) .

From a chemistry point of view the 6s2 inert pairs in mercury are fluctuating as Cooper pairs in the superconducting state below 4K , and as localized , valence shell electrons , spherically distributed around the Hg2+ kernel , above 4K . Then they are tightly bonded by the kernels , but below 4K they are released from them and are completely free in the interatomic void spaces . I speculated that this might be caused by a redox effect . The standard reduction potential of Hg2+ at STP is 0.851 V versus the SHE . Undoubtedly this potential falls with a falling temperature of the mercury cations . Perhaps at 4K E0red = 0 V , and the Hg2+ kernels , which are strongly oxidizing and electrophilic at room temperature , no longer are . They lose their electrophilic grip on the inert pairs , which float off into the lattice voids as Cooper pairs . They have been freed from their parent kernels , and are now free electrons , just like the static electricity electrons in lightning bolts and Tesla coil sparks .

As the mercury is warmed above 4K , the Hg2+ kernels become electrophilic again and can quickly recapture their itinerant 6s2 inert pairs . Suppose a chemical compound containing Hg0 was designed so that the Hg2+ kernels were prevented from recovering the inert pairs . Then the electron pairs would stay in the interatomic voids at higher temperatures , possibly even at room temperature . The inert pairs would also have a tendency to stay together as pairs , because they are still coulombically attracted to the underlying Hg2+ parent kernels at all temperatures , even though they are no longer bonded to them . The Hg0 compound might thus be a high temperature superconductor , possibly even a room temperature superconductor !

The “chemical trick” I suggested that might accomplish this objective was to include the Hg0 in a suitable sulfide “crystal container”. Sulfide anion is a natural reducer and will donate charge to the Hg2+ kernels when bonded to them :

Hg2+ + 2e ----> Hg0 ; E0red = 0.851 V ;

S2- – 2e -------> S0 ; E0ox = 0.476 V ;

Net reaction : Hg2+ + S2- -------> Hg0–S0 ; E0T = 1.327 V .

The high cell potential E0T for the overall reaction indicates it would be thermodynamically favorable – very much so – at STP . Note that the mercury and sulfur don't actually separate out as the elements ; rather , the redox equation above suggests that very strong S>Hg coordinate covalent bonds are formed in the Hg2+–S2- compounds .

Hg(II) demonstrates a strong preference for a two-fold linear sp coordination by anions and ligands , as in the mineral cinnabar , a-HgS (red) , with linear S–Hg–S bonds ; it also prefers the four-fold tetrahedral sp3 coordination , as in metacinnabar , b-HgS (black) , which has the cubic zincblende crystal structure . Two possible “crystal containers” for tetrahedrally-coordinated Hg0 were considered : the wurtzite–zincblende (sulfides) and the thiospinel structures .

Synthesis of the proposed wurtzite–zincblendes and thiospinels containing Hg0 could probably be achieved only in a high pressure , high temperature (HP–HT) apparatus , such as a tetrahedral anvil or belt type of press . High pressure would force the sulfide anions down onto the mercury atoms , “popping” the 6s2 inert pairs off their Hg2+ kernels and into the interatomic void space . Clustering of the sulfide anions around each Hg2+ kernel in a tetrahedral geometry , forming the strong S>Hg coordinate covalent bonds , will

* energetically assist the popping of the inert pairs into the void spaces ;

* occupy the 6s,p Hg orbitals with the sp3 hybrid orbital used for the S>Hg coordinate covalent bonds (or , alternately , sterically block them) , thereby preventing the inert pairs from infiltrating back onto the Hg2+ kernels when the compound is cooled back to room temperature and depressurized ;

* coulombically repel the inert pair electrons from the Hg2+ kernel , forcing the electrons to “keep their distance” from it ; and ,

* electronically “neutralize” the strongly oxidizing and electrophilic Hg2+ kernel , effectively cancelling its oxidizing and electrophilic nature .

A variety of wurtzite–zincblendes and thiospinels containing Hg0 were surveyed in the Inert Pairs web page and continued in the Thiospinels and Mixed-Valent Inert Pair Systems essays . Compounds containing other heavy metal inert pairs such as Tl(I) , Sn(II) , Pb(II) , and Bi(III) were also proposed for study in those web pages . I was in the middle of a fourth web page on the subject of inert pair superconductors , exploring the design and synthesis of covalent–metallic compounds combining heavy metal inert pair elements with sulfides of Family IIIA/13 and VA/15 , when an exciting new design concept occurred to me .

I was studying the possibility of using the double wurtzite (or zinc blende) structure as the crystal container for the heavy metal elements with the inert pairs . For example , Hg0 could be triturated with an equimolar quantity of the substrate boron sulfide , B2S3 [GIF image , 70 KB] , to form a “premix” in which the mercury atoms are weakly bonded to the sulfurs . This premix material would then be “cooked” in a press under HP–HT conditions to form the covalent–metallic compound Hg0B2S3 , in which mercury's 6s2 inert pairs have been “popped” into the interatomic void space : (Hg2+)B2S3(2e2-) :

Hg0 (b.p. 357 C) + B2S3 (m.p. 563 C) -------- [triturate] -------- Hg0B2S3 premix

-------- [anvil press , HPHT] -------> Hg0B2S3 double wurtzite (or zinc blende) .

All the atoms in the compound should be tetrahedrally coordinated with their neighbours , so the material would have the double wurtzite (or zinc blende) crystal structure with alternating layers of Hg–S and B–S . A careful electron count [GIF image , 45 KB] confirmed that each atom would have an octet of valence shell electrons , with two electrons – formally the Hg inert pair – left over as “extra” in the lattice . The A0M2S3 compounds (A = Zn , Cd , and Hg ; M = B , Al , P , As , Sb , and Bi) would be interesting materials to synthesize and study . I'll mention them again below , but in a different and much more interesting form !


The Tetrahedral Voids Design Concept


I had known of another double zinc blende , Cu2HgI4 , for a long time . It's interesting in two ways : first , it's thermochromic , being brick red at room temperature and turning black (or maybe “chocolate-brown”) at ~ 70 C . Second , it's an ionic conductor ; at higher temperatures the Cu1+ cations can move through the lattice under an applied potential difference . They can do this because of the presence of empty tetrahedral voids , in stoichiometric numbers (that is , not defects) , as part of the compound's empirical formula : Cu2Hg[ ]I4 , where [ ] denotes the empty void space . The voids are associated with the Hg2+ cations : the red , room temperature , tetragonal zinc blende form is made up of alternating layers of Hg2+–I1-–[ ]–I1- and Cu1+–I1-–Cu1+–I1- . In the high temperature black form (having the cubic zincblende structure) the cations and void spaces are randomly distributed in the lattice . At elevated temperatures the Cu1+ cations can be drawn through the crystal , via the empty voids , under an applied p.d. :

The above molecular model was based on a sketch of Cu2Hg[ ]I4 in the student experiment , “Thermochromism in the Ionic Conductor , Cu2HgI4(PDF , 223 KB) . The sketch in that Web document was in turn copied from a chemistry textbook by A.B. Ellis et al. , Teaching General Chemistry , A Materials Science Companion , American Chemical Society , Washington , D.C. , 1993 ; I used to have a copy of Ellis's book a long time ago .

The synthesis of Cu2Hg[ ]I4 has been described in several student experiments . A procedure for it is included in Walton's lab manual , which I used in my second year college inorganic chemistry course : H.F. Walton , Inorganic Preparations , A Laboratory Manual , Prentice-Hall , New York , 1948 ; pp. 79-81 . This useful inorganic synthesis textbook can be downloaded for free from the Sciencemadness.org library resources web page [DJVU , 1644 KB ; a suitable DjVu reader for your computer can be downloaded for free from djvu.org . The WinDjView Reader v. 1.0.3 for older FAT32 Windows can be downloaded for free from FileHorse] . Two methods of preparing Cu2Hg[ ]I4 have also been described in Brauer's compendium : G. Brauer (ed.) , Handbook of Preparative Inorganic Chemistry , Vol. 1 , 2nd edition , Academic Press , New York , 1963 ; pp. 1110-1111 . This immense inorganic chemistry synthesis reference (Vols. 1 & 2 combined) can be downloaded for free from Sciencemadness.org [PDF , 19,090 KB] . Note : this PDF file can be opened only with Adobe Acrobat Reader v. 6 or later . If desired , a suitable version of this application can be downloaded for free from Oldversion.com .

While my class never carried out the preparation of Cu2Hg[ ]I4 described in Walton's lab manual , I was certainly aware of the compound at that time (late 1960s) . My interest in it was revived when I found the Web document of the University of Massachusetts student experiment cited above .

As I was musing over the proximity of the Hg2+ and empty tetrahedral voids in Cu2Hg[ ]I4 the idea suddenly came to me : suppose layers of Hg0 and [ ] were placed together in a zinc blende , and the material was subjected to HPHT conditions ; would it be possible to pop the mercury atoms' 6s2 inert pairs straight into the tetrahedral voids ? This should be energetically feasible because they are enormous volumes compared to the tightly restrictive interatomic void space that would have to be used by any popped inert pairs in the superconductor candidate compounds discussed earlier , such as Hg0B2S3 .

Since the Cu(I) cations are mobile through the voids in Cu2Hg[ ]I4 at higher temperatures , surely the pairs of electrons should similarly be very mobile in them . When the electron pairs are in those voids , they will have completely escaped from the underlying Hg2+ cations , which are themselves strongly bonded to and “pacified” by the tetrahedrally-surrounding iodide anions . The freed electron pairs in the voids will no longer be subject to the Fermi-Dirac distribution as are the associated electrons of the metallic bond XO in conventional metallic solids . They will be completely free ; they should be able to flow unimpeded , unopposed , with no resistance , throughout the lattice in the void channels .

The concept of electrons being the charge and energy carriers in metals was first proposed in the electron theory of metals (1900) by the German physicist Paul Drude (1863-1906) :

Drude pictured the mobile , free electrons in the metallic bond as behaving in a manner similar to monoatomic gas molecules , following Maxwell-Boltzmann statistics . His theory explained the physical properties of metals fairly well (he was primarily interested in their optical properties) . The rather glaring discrepancy between his theoretical and the actual values of the heat capacities of the common metals was resolved in the late 1920s by a modification , based on a quantum mechanical treatment , of his original theory . The quantum modification was the requirement that the electrons in a metal were associated and therefore must follow Fermi-Dirac statistics , as mentioned above .

However , I think Drude may have still been at least partially correct , as the Cooper pairs in the superconducting state in metallic solids may be free in nature and therefore resemble the kind of perfect gas molecules envisioned by him . As this web page unfolds (or scrolls downward) I'll discuss compounds containing pairs of electrons that are similarly unassociated and that should result in superconductivity in the material concerned . I would propose that these free electrons inside matter could be called Drude electron gas , while the associated electrons in conventional metallic solids could be referred to as Fermi-Dirac electron gas . The free electrons outside matter are , of course , the various manifestations of static electricity .

Another advantage of locating the electron pairs in the voids is that they will be “focused” by the lone pairs on the anions that tetrahedrally surround them . The negative lone pairs will act like tiny tetrahedral pistons in a press to squeeze the free electrons and suspend them precisely in the centers of the voids :

As they are drawn downfield in an applied p.d. the electron pairs will stay precisely in the void centers as they hop simultaneously from void to void , thus avoiding any violent impacts on the anions that surround the voids . This will help to maintain their pairing integrity at very high temperatures .

Another consideration applying to the electron pairs in the voids is the strength of their pairing . They will have to be in an antiparallel ordering in the void , and they will be physically very close together . That is , their coherence lengths will be extremely short , the shortest possible that can be achieved in any superconductor . According to the BCS theory this very short coherence length will result in an exceptionally high transition temperature for the material . I had previously derived a simple analysis of the magnetic to electric forces between two antiparallel (or conversely , parallel) electrons in a Cooper pair (GIF image , 34 KB) . If they are in a perfectly antiparallel orientation the magnetic force binding them together is ~ 1200 (or maybe even ~ 2400 , GIF image , 12 KB) times as strong as the electric force that would repel them apart . Also , the magnetic force is inversely proportional to the square of the separation distance of the two electrons . Since the two electrons are so close together in the voids , the binding magnetic force between them should be quite substantial . As a result the material could be a fully functional ambient superconductor and at even higher temperatures , possibly up to its melting or decomposition point . As long as the lattice stays intact with the voids still in place , and the electrons remain paired in the voids , the solid will continue to be superconducting .

I should also point out that the space within the voids is a perfect vacuum , with no gas molecules or atoms to disrupt the flow of the electrons . The vacuums in the old radio vacuum tubes , fluorescent light bulbs , cathode ray tubes (my computer monitor !) , and in the colorful neon tubes in commercial displays are all very inefficient compared to the vacuum in the void channels . It's an even better vacuum than that of interstellar space ! It's near to being an absolute vacuum . That might explain the phenomenon of the superconducting current in a ring-shaped conductor that will circulate forever around the ring , once set in motion . After the electrons have been given an intial push (momentum) , they will move through the interatomic voids forever , with nothing whatsoever to stop their passage . There's no resistance to their motion , because there are no atoms in the void space against which they can impact .

Note that this ring current phenomenon has been observed in the classic BCS superconductors which require refrigeration in liquid helium at 4 K . At that temperature the volatility of the atoms in the lattice is essentially nil . As these new void-type materials are warmed up some of the lattice atoms , or volatile impurity atoms trapped in the lattice , might diffuse into the voids , thereby blocking the channels and preventing the passage of the electrons . Thermal diffusion of stray atoms could thus conceivably degrade the superconducting properties of the void compounds at higher temperatures .

In the free electron pair candidate compounds without the large void spaces such as Hg0B2S3 , the vibrating atomic kernels (phonons) could block the motion of the electrons through the lattice . However , the phonons are less likely to disrupt the passage of the electrons flowing in the large void channels , which are vastly larger than ordinary interatomic void space . And as pointed out , the anion stereochemical lone pairs of electrons pointing into the voids will suspend the electrons in their centers , helping to prevent their impact on the anions .

Zinc blendes have the general formula MX , where M is a metal (or metalloid) atom – formally a cation – and X is a nonmetal atom , formally an anion . A zinc blende like Cu2Hg[*]I4 has four atoms of each type , with the empirical formula M4X4 . The chemistry of the material may result in a missing atom in the formula ; this results in a missing atom , ie. a tetrahedral void , in the lattice of the solid state material . For example , Cu2Hg[ ]I4 is comprised of 2 CuI + HgI2 , and in fact it can be synthesized by melting together those reagents in that 2 : 1 stoichiometric ratio (Brauer) . To mentally assist in designing new zinc blende compounds I sketched out a picture symbolically representing them as such :

The M0[ ]Si2S4 compounds (M = Zn , Cd , and Hg) should be isostructural with Cu2Hg[ ]I4 :

Silicon disulfide was chosen as the host for the M0, since the silicon(IV) atoms always have a tetrahedral coordination by the sulfurs , even under very high pressure , as was demonstrated by the determination by Prewitt and Young (1965 , DOI) of the crystal structure of SiS2 in a HPHT synthesis :

Boron(III) is also guaranteed to be tetrahedrally coordinated by sulfur atoms in these zinc blendes . The base kernels of the heavy metal elements selected as donors of the inert pairs hopefully will also accept a tetrahedral coordination by the sulfurs . In fact Zn2+, Cd2+, and Hg2+ all strongly favor such a tetrahedral coordination , as in the zincblende compounds ZnS , CdS , and b-HgS (the black sulfide of mercury ; ZnS and CdS also have wurtzite crystal structures) . I'm not as certain of the situation with the Tl(III) , Pb(IV) , and Bi(V) base kernels , although there's no theoretical prohibition for them not to be tetrahedrally coordinated in zinc blendes .

Mercury(0) silicon disulfide might be prepared by the insertion of Hg0 into the host substrate silicon disulfide (GIF image , 55 KB) in a 1 : 2 molar ratio to form a sort of premix” ; this might be accomplished by the gentle trituration of the mercury metal and SiS2 in a mortar . The premix material would then be converted into the covalentmetallic zinc blende in a press under HPHT conditions :

Hg0 (b.p. 357 C) + 2 SiS2 (m.p. 1090 C) -------- [triturate in mortar] ------->

Hg0Si2S4 premix-------- [HPHT] -------> Hg0[ ]Si2S4 = Hg2+[**]Si2S4 .

Alternately , silicon disulfide might be formed in situ by the combination of silicon powder and flowers of sulfur under HPHT conditions , as described by Prewitt and Young (graphic above) . Black mercuric sulfide , in which the Hg(II) is already tetrahedrally coordinated by the sulfides , would be added to the silicon and sulfur reagents , and the mixture thoroughly ground together in a mortar . This mixture would then be cooked” in the press :

b-HgS (black , m.p. 850 C) + 2 Si0 (m.p. 1414 C) + 3 S0 (m.p. 115 C , b.p. 445 C)

-------- [HPHT] -------> Hg0[ ]Si2S4 = Hg2+[**]Si2S4 .

This latter procedure might be more convenient than the first one , as the solid HgS would be easier to manipulate than the volatile , toxic liquid mercury . It would also be a more economical method , as silicon and sulfur are much cheaper than the somewhat obscure reagent silicon disulfide (which , incidently , is quite sensitive to atmospheric humidity to the point where it has been described as smelling like rotten eggs from the H2S evolving from its surface) .

Note that silicon (or its derivative ferrosilicon) reduces the Mg2+ in magnesium oxide to elementary magnesium (vapor) in the Pidgeon Process , a commercial production method for the manufacture of magnesium metal . The silicon in the above formula should therefore have no problem in reducing the Hg(II) in HgS to Hg0 [in the formal sense , since the mercury will remain Hg(II) , coordinated by the sulfurs , while the reduction electrons will enter the tetrahedral voids as the Cooper pairs] .

In a second experimental note , silicon powder will react with sulfur at ~ 500–700 C to form the SiS2 component , for example as described in European Patent Application 0802161 A1 (K. Yamamoto and N. Ikeda , 1997) , “Method of Manufacturing Silicon Sulfide” (PDF , 248 KB) :

Yamamoto and Ikeda stressed the importance of coating the fine silicon particles with molten sulfur , as untreated silicon powder tended to sinter at those elevated temperatures , reducing its reactivity with the sulfur . However , I think the simple solution to the sintering problem would be to grind the silicon , sulfur , and mercury sulfide components , weighed carefully in the correct stoichiometric proportions , in a mortar . The thoroughly ground mixture would be placed in the compression capsule or chamber of the press and submitted to the HP–HT synthesis .

The HPHT reaction outlined above could probably be carried out in this lower 500–700 C temperature range , and not at 1300 C as was used by Prewitt and Young in their preparation of SiS2 .

The analogous zinc compound would be the most economical of all of the zinc blendes , since its component elements Zn , Si , and S are very abundant and cheap :

ZnS (m.p. 1700 C) + 2 Si0 (m.p. 1414 C) + 3 S0 (m.p. 115 C , b.p. 445 C)

-------- [HPHT] -------> Zn0[ ]Si2S4 = Zn2+[**]Si2S4 .

Zinc metal , silicon , and sulfur are also innocuous , relatively non-toxic chemical reagents .

Caution : zinc powder combines with flowers of sulfur in a violent , very exothermic Thermit-like reaction , so ZnS is recommended as the source of the zinc atoms in Zn2+[**]Si2S4 . See these impressive YouTube videos for demonstrations of the zinc–sulfur reaction : this demonstration from the University of Nottingham , UK [MP4 , 13,485 KB , runtime 3:10] ; this one showing a Zn + S mixture being used as a rocket propellant [MP4 , 731 KB , runtime 0:09] ; and this one illustrating the fiery combustion of Zn + S [MP4 , 1621 KB , runtime 0:19] .

The use of tin(0) as the free electron pair source in SnIV[**]B2S4 is also quite interesting :

Sn0 (m.p. 232 C) + S0 (m.p. 115 C , b.p. 445 C) + B2S3 (m.p. 563 C)

-------- [HPHT] -------> SnIV[**]B2S4 .

The components are all fairly low melting solids , and are innocuous , readily available , inexpensive reagents . As a further variation , the B(III) might be replaced by a combination of Zn(II) and Si(IV) :

Sn0 + Si0 (m.p. 1414 C) + ZnS (m.p. 1700 C) + 3 S0 ------- [HPHT] ------> SnIV[**]SiZnS4 .

Zinc and cadmium are natural reducing agents , and their corresponding Zn2+ and Cd2+ cations are low energy , redox-wise (neither oxidizing nor reducing) , so they have no inclination to try to recover their popped 4s2 and 5s2 valence shell electrons , respectively , from the tetrahedral voids :

Zn2+ + 2e ----> Zn0 ; E0red = – 0.7618 V ;

Cd2+ + 2e ----> Cd0 ; E0red = – 0.403 V .

This means it's unnecessary to use a naturally reducing anion like sulfide to electronically neutralize the Zn2+ and Cd2+ ; they are already electronically innocuous . In fact , as the two redox half equations above tell us , it would actually be energetically favourable for the Zn0 and Cd0 to transfer their 4s2 and 5s2 valence shell electrons into the voids . It should thus be possible to prepare zinc and cadmium zinc blendes (but not those of mercury , which must be stabilized by charge transfer with sulfide anions) with electron pairs in their tetrahedral voids , using oxide anions . The following zinc and cadmium oxide compounds would be quite interesting to prepare and study :

Zn0 (m.p. 420 C , b.p. 907 C) + 2 SiO2 (m.p. 1722 C)

-------- [HPHT] -------> Zn2+[**]Si2O4 ;

Cd0 (m.p. 321 C , b.p. 767 C) + 2 SiO2 -------- [HPHT] -------> Cd2+[**]Si2O4 ;

ZnO (m.p. 1974 C) + Si0 (m.p. 1414 C) + 1 SiO2 -------- [HPHT] -------> Zn2+[**]Si2O4 ;

CdO (m.p. 1559 C , subl.) + Si0 + 1 SiO2 -------- [HPHT] -------> Cd2+[**]Si2O4 .

As mentioned above , silicon is a remarkably powerful reducing agent , and can reduce Mg2+ to Mg0 in the Pidgeon Process , so it should be able to easily reduce Zn2+ and Cd2+ to Zn0 and Cd0 (in the formal sense , since the reduction electrons will actually go into the tetrahedral voids) . The compound silicon monoxide , SiO , is well known and is commercially available (eg. Alfa-Aesar) at a modest cost . It could also act as the reducing agent in the formation of the zinc blendes :

ZnO + SiO + SiO2 -------- [HPHT] -------> Zn2+[**]Si2O4 ; and ,

CdO + SiO + SiO2 -------- [HPHT] -------> Cd2+[**]Si2O4 .

Silica and zinc metal are very abundant , inexpensive commodities ,

and are relatively non-toxic , while cadmium is a less common element and is considered to be a toxic heavy metal . Also , zinc oxide has the wurtzite crystal structure , which is encouraging ; cadmium oxide , however , has a cubic rocksalt structure , with octahedral not tetrahedral Cd2+. Zn2+[**]Si2O4 , if it proved to be a zinc blende or a wurtzite with tetrahedral voids , and if it was a high temperature (even ambient) superconductor , could be an extremely important and valuable material in the technology of the future .

Now we know the “chemical trick” for doing so , Drude metal triple zinc blendes can be designed using the oxides and sulfides of the IIIA/13 and VA/15 families of elements , as was discussed earlier in this web page . For example , the Hg0B2S3 combination can be reformulated using a 1 : 2 molar ratio of guest atoms to host lattice to accomplish this objective :

Hg0 (b.p. 357 C) + 2 B2S3 (m.p. 563 C) -------- [triturate] -------- Hg0B4S6 premix

-------- [anvil press , HPHT] -------> Hg0[ ]B4S6 = Hg2+[**]B4S6 = Hg2+B2S3[**]B2S3 .

Oxide host lattices could be employed for guest atoms which are natural reducing agents (for example , zinc and cadmium) in analogous zinc blende syntheses :

Zn0 (m.p. 420 C) + 2 B2O3 (m.p. 450 C) -------- [heat , ambient pressure] ------->

Zn0[ ]B4O6 = Zn2+[**]B4O6 = Zn2+B2O3[**]B2O3 .

This reaction could be tried for several of the M(III) oxides and sulfides , as host lattices , of the IIIA/13 and VA/15 families of elements (M = B , Al , P , As , Sb , and Bi) . Many naturally reducing metal elements could theoretically dissolve in these oxides and sulfides to form zinc blendes which would be excellent superconductor candidate compounds .




The F-centers , or “color centers”, are an example of electrons trapped in the cavities of crystals . When a colorless , transparent crystal of sodium chloride is irradiated with a narrow beam of high energy X-rays , a circular , intensely reddish-orange patch will form in the irradiated section . Similarly , the irradiation of a crystal of potassium chloride by high energy X-rays will produce a violet colored F-center in it . The X-rays have knocked electrons off some of the chloride anions ; they become chlorine atoms which can migrate and diffuse out of the crystal lattice . The displaced electrons can resonate over the surrounding sodium cations in the empty octahedral cavities vacated by the chlorine atoms .

In Van Nostrand’s Scientific Encyclopedia , 4th edition , 1968 , p. 374 , there is a nice color photo of such irradiated salt crystals . The F-centers are described as follows : “Certain crystals , such as the alkali halides , can be colored by the introduction of excess alkali metal into the lattice , or by irradiation with X-rays , energetic electrons , etc. Thus sodium chloride acquires a yellow color and potassium chloride a blue-violet color” (p. 384) . F-centers are also discussed by Moore , Seven Solid States (cited above) , pp. 37-39 .

The F-centers represent a sort of mixed-valent Na0–Na1+ compound , with the free electrons resonating over the base of sodium cations surrounding the cavities , possibly in the sodiums' 3s orbitals . The free electron resonance in the octahedral anion cavities can absorb certain wavelengths of light passing through the crystals , resulting in the strong coloring of the F-centers . However , these mixed-valent zones are relatively dilute in the crystal matrix ; they certainly aren't stoichiometric M0–M1+ in the entire mass of the crystal . The F-center compound would be a metallic solid if the electron–cavities were present in stoichiometric quantities in it : Na1+[(Cl1-)x(e1-)1-x] , x = 0.5 ; that is , in the hypothetical cubic rocksalt compound Na2[*]Cl with octahedral cavities containing the electrons .

Unlike the F-centers' resonating electrons ,

* Electrons in the wurtzites' tetrahedral voids would be in cation vacancies , unlike those in the F-centers' anion (halide) vacancies ;

* Electrons in NaCl F-centers are associated with the surrounding sodium cations , and resonate in their 3s frontier orbitals ; the free electrons in the wurtzite tetrahedral voids are unassociated with their parent metal cation kernels ;

* The wurtzite electron–voids would be in stoichiometric quantities in their empirical formulas , compared to the very dilute electron–voids in the F-centers ; and ,

* As mentioned , the mobility of the Cu1+ cations in the tetrahedral voids of the ionic conductor Cu2HgI4 strongly implies that the electrons in the voids would similarly be highly mobile . While being able to resonate in their cavities , the F-center electrons are completely localized in them .

The proposed double wurtzites with electrons trapped in tetrahedral voids would essentially be a new type of electride . An electride is a material in which electrons have replaced anions in the lattice (as in the F-centers' Na1+[(Cl1-)x(e1-)1-x] compound above) . The ammonia electrides with the Alkali and Alkaline Earth elements are a chemical system in which free electrons are thought to be trapped in intermolecular voids in the ammonia solutions and in frozen ammonia . In fact , pairs of such electrons , loosely associated with the ammoniated metal cations , have been hypothesized in the electrides . Such electron pairs were first suggested in 1945–46 research reports by the American physicist R.A. Ogg , who studied the electronic properties of the electrides :

“....... two electrons trapped in the same cavity are appreciably stable with respect to either two electrons trapped in separate cavities or one trapped and one conducting electron” (p. 155)

– R.A. Ogg Jr. , “Electronic Processes in Liquid Dielectric Media . The Properties of Metal–Ammonia Solutions”, J. Amer. Chem. Soc. 68 (1) , p. 155 (1946) [DOI] ; several more of Ogg's research papers about the ammonia electrides :

“The Conduction Process in Dilute Metal–Ammonia Solutions”, J. Chem. Phys. 13 (11) , p. 533 (1945) [DOI] ; “The Absorption Spectrum of Metal–Ammonia Solutions”, J. Chem. Phys. 14 (2) , pp. 114-115 (1946) [DOI] ; “Bose–Einstein Condensation of Trapped Electron Pairs . Phase Separation and Superconductivity of Metal–Ammonia Solutions”, Phys. Rev. 69 (5&6) , pp.243-244 (1946) [DOI] ; “Superconductivity in Solid Metal–Ammonia Solutions”, Phys. Rev. 70 (1&2) , p. 93 (1946) [DOI] .

Ogg claimed to have detected superconductivity in frozen ammonia electride solutions , with “persistent currents” measured up to the melting point of ammonia , at –77 C , or 196 K . Unfortunately his results couldn't be reproduced , at least by one team of researchers :

H.A. Boorse et al. , “The Electrical Conductivity of Rapidly Frozen Solutions of Sodium in Liquid Ammonia”, Phys. Rev. 70 (1&2) , pp. 92-93 (1946) [DOI] .

Failure by other researchers to detect any superconductivity in the ammonia electrides resulted in Ogg's findings to be rejected as an experimental artifact or to be entirely discredited . Nevertheless the existence of “solvated electrons” and electron pairs in the ammonia electrides is now considered to be an established fact and orthodox science . Various reviews of electride chemistry :

J.L. Dye , “The Solvated Electron”, Scientific American 216 (2) , pp. 76-83 (February , 1967) [nice photos , DOI] ; J.L. Dye , “Electrides , Negatively Charged Metal Ions , and Related Phenomena”, Prog. Inorg. Chem. 32 , pp. 327-441 , S.J. Lippard (ed.) , John Wiley , New York , 1984 [DOI] ; J.L. Dye , “Electrides : Ionic Salts with Electrons as the Anions”, Science 247 (4943) , pp. 663-668 (1990) [DOI] ; M.J. Wagner and J.L. Dye , “Alkalides , Electrides , and Expanded Metals”, Ann. Rev. Mater. Sci. 23 , pp. 223-253 , R.A. Huggins et al. (eds.) , Annual Reviews , Palo Alto , CA , 1993 [DOI] ; P.P. Edwards , “The Electronic Properties of Metal Solutions in Liquid Ammonia and Related Solvents”, Adv. Inorg. Chem. Radiochem. 25 , pp. 135-185 , H.J. Emelus and H.G. Sharpe (eds.) , Academic Press , New York , 1982 [DOI] ; W.L. Jolly , “Metal-Ammonia Solutions”, Prog. Inorg. Chem. 1 , pp. 235-281 , F.A. Cotton (ed.) , Interscience , New York , 1959 [DOI] ; M.C.R. Symons , “Nature of Metal Solutions”, Quart. Rev. 13 (2) , pp. 99-115 (1959) [DOI] ; M.C.R. Symons , “Solutions of Metals : Solvated Electrons”, Chem. Soc. Rev. 5 (4) , pp. 337-358 (1976) [DOI] ; J.L. Dye , “Electrides : Early Examples of Quantum Confinement”, Accts. Chem. Res. 42 (10) , pp. 1564-1572 (2009) [DOI] ; A. Torrisi , “Electrides : Electrons Claim Their Independence”, Opticon1826 (11) , 5 pp. (Autumn , 2011) [PDF , 343 KB] .

A demonstration of the solution of a lump of sodium metal in liquid ammonia is provided in the YouTube video “Sodium Electride” [MP4 , 1991 KB , runtime 0:43] . Another YouTube video , “Solvation of Electrons”, shows the solution of piece of lithium foil in home-made liquid ammonia [MP4 , 8231 KB , runtime 2:18] .

Solid state electrides can be produced by the insertion of electrons into the centers of spherical zeolite cages” :

The above drawing of the C12A7:e zeolite , with the formula [Ca24Al28O68]4+ 4e , was copied from the report by H. Buchammagari et al. (Professor H. Hosono's group) , “Room Temperature-Stable Electride as a Synthetic Organic Reagent : Application to Pinacol Coupling Reaction in Aqueous Media”, Org. Lett. 9 (21) , pp. 4287-4289 (2007) [DOI] ; Fig. 1 , p. 4287 . My thanks to the author and/or copyright holder of the original sketch . By the way , the trapped electrons are represented by the large green spheres in the drawing .

A metallic zeolite electride has even been found to be superconducting , although only very near to Absolute Zero (Tc ~ 0.4 K) : M. Miyakawa et al. , “Superconductivity in an Inorganic Electride 12CaO.7Al2O3 : e ”, J. Amer. Chem. Soc. 129 (23) , pp. 7270-7271 (2007) [DOI] . As you can see in the above sketch , the trapped electrons in the zeolite spheres don't have any sort of channels by which they can escape from their cages” and flow through the lattice under an applied potential difference . They are all dressed up with nowhere to go”, to use a popular expression . It's really quite remarkable that the zeolite electrides are metallic at all , let alone superconducting .

In contrast ,

* the zinc blende electrons are in a sponge-like network of channels through which they can pour through the lattice ;

* they are in a much greater density in the crystal than are the electrons in the zeolites ;

* they are in pairs , which probably resemble Cooper pairs ;

* the zinc blendes are chemically and crystallographically much simpler and more compact than the zeolites ;

* they are much more atom-efficient” than the zeolites , less amounts of chemical reagents being used to produce more void electrons .

The zinc blendes (or wurtzites) with void electrons could reasonably be considered as super-electrides , with the potential of providing an entirely new generation of materials with extraordinary electronic properties . They have the potential to take electride chemistry to a stratospheric performance level unattainable by existing electrides .


Design and Synthesis of Zinc Blende Iodides with Tetrahedral Voids


Mercury(II) bonds strongly with iodide anions in the very stable tetrahedral complex [HgI4]2-, as in the ionic conductor Cu2Hg[ ]I4 mentioned above . The iodides form coordinate covalent bonds with the “empty” Hg2+ cation , using the latter's 6s + 6px,y,z orbitals , hybridized to the tetrahedral sp3 orbital . Iodide anion can act as an “unnatural” reducing agent (with a negative oxidation potential) , being forced by the strongly oxidizing Hg2+ cation to transfer electrons to it :

2 I1- – 2e -------> 2 I0 ; E0ox = – 0.5355 V ;

Hg2+ + 2e ----> Hg0 ; E0red = 0.851 V ;

Net reaction : Hg2+ + 2 I1- -------> I0–Hg0–I0 ; E0T = 0.3155 V .

When Hg2+ and 2 I1- are mixed together elemental mercury and iodine don't separate out ; instead , the insoluble (0.06 g/L) compound HgI2 precipitates from solution . The positive cell potential E0T for the above redox reaction does indicate , however , that strong Hg–I coordinate covalent bonds form in the HgI2 , with some I–>Hg charge transfer .

Zinc blende (or wurtzite) iodides with crystal structures analogous to that of Cu2Hg[ ]I4 can be designed in the same way as the M2+[**]Si2S4 series , as shown in the following sketch :

The hypothetical double wurtzite Hg2+[**]Zn2I4 should be isostructural with Cu2Hg[ ]I4 , which can be written as Hg2+[ ]Cu2I4 . However , in the former compound mercury's 6s2 inert pairs will be located in the tetrahedral void spaces , which are vacant in Cu2Hg[ ]I4 .

Hg2+[**]Zn2I4 might be synthesized simply by the insertion of one equivalent of Hg0 into two equivalents of ZnI2 under HPHT conditions :

Hg0 (b.p. 357 C) + 2 ZnI2 (m.p. 446 C , b.p. 625 C , dec.)

-------- [HPHT] -------> Hg2+[**]Zn2I4 .

The low melting point of ZnI2 and its high solubility in water and organic solvents suggest that its Zn–I bonds are predominantly covalent (or coordinate covalent) . Zinc iodide is a readily available , relatively inexpensive reagent . The synthesis of ZnI2 from its component elements using an organic solvent such as methanol (PDF , 172 KB) or ethanol (web page) has been described in a student experiment . Zinc iodide is very hygroscopic , and should be dried in a drying oven at 110 C for several hours , then cooled to room temperature in a dessicator before use .

The bismuth compound , BiLiI4 , seems to have two tetrahedral voids . One of them might be filled with a Hg0 atom (to maintain electrical neutrality in the lattice) . Then , two electron pairs per formula unit could be produced in the material , one of which would be located in a tetrahedral cation vacancy ; the other would presumably be in the interatomic void space .


High Pressure Forms of Bismuth(III) Iodide and Lead(II) Iodide


Bismuth(III) iodide has a relatively open, low dimensional crystal structure :

This sketch was copied from the Wikipedia web page , Bismuth(III) iodide . Again , my thanks to the author of this sketch , and Wikipedia , for implied permission to reproduce it on this web page .

Note carefully that the Bi(III) in this structure has a symmetrical , undistorted octahedral coordination by the neighbouring iodides . This is unusual , since Bi(III) is 6s2 electronically . Where are those 6s2 inert pairs ? A possible answer is that they are unhybridized , and are surrounding the Bi(V) kernel in the spherical 6s orbital . The physical properties of BiI3 , such as its crystal structure shown above and its low melting point (409 C) , imply that its BiI bonds are coordinate covalent , not ionic in nature . A possible bonding scheme for BiI3 with these considerations in mind is presented in the following sketch :

The reader is , of course , familiar with the concepts of innner and outer octahedral coordinate covalent bonding , for example in the iron(II) compounds potassium ferrocyanide , K4Fe(CN)6 (inner octahedral , d2sp3 hybridization , low spin Fe2+, diamagnetic) , and in K4FeF6 (outer octahedral , sp3d2 hybridization , high spin Fe2+, strongly paramagnetic) . In the former complex the nucleophilic cyanide anions bond strongly with the Fe(II) and push deeper into its kernel ; in the latter compound , K4FeF6 , the fluoride anions (which resemble non-nucleophilic little teflon balls) are feebly bonded to the iron kernels and don't push very deeply into them .

The bonding situation in BiI3 is analogous to these iron(II) compounds . Under mild ambient conditions the iodides , while forming fairly strong coordinate covalent BiI bonds , can't push very far into the kernel , which remains Bi(III) , surrounded by the unhybridized 6s2 inert pair . Bismuth is obliged to use the outer octahedral sp5 hybrid orbital to form its BiI bonds .

Suppose a sample of BiI3 is subjected to HPHT conditions in a press . Its atoms might be compressed into a more compact crystal structure . Compressed BiI3 could conceivably adopt the simple cubic rhenium trioxide structure , in which the Bi(III) is using an inner octahedral sp5 hybrid orbital for its BiI bonds :

In order to form such an inner octahedral hybrid orbital the 6s2 inert pair must be popped out of the 6s orbital and into the large central , empty cavity of the ReO3 structure :

The inert pair electrons in the cavities will be surrounded by twelve lone pairs of electrons on the iodide anions (5s2 5p6) , which should have a linear sp hybridization : 5spx4 + 5py2 + 5px2 . In perovskites the central cavity of the ReO3 structure is occupied by large monovalent , divalent , and trivalent cations , so there would undoubtedly be more than enough room in the central voids for the popped electron pairs . Such a symmetrical , isotropic , cubic rhenium trioxide crystal structure for Bi[**]I3 would have a three dimensional network of void channels that would be the perfect pathways for the electron pairs to move through the lattice . The iodide lone pairs that surround the voids would keep the electron pairs suspended in the void centers , helping them to avoid impacts on the atomic kernels as they hop downfield under the applied potential difference .

When compressed , BiI3 might recrystallize into the more compact ReO3 structure , but the Bi hybridization could be either outer octahedral , resulting in a non-metallic form with localized 6s2 inert pairs , or the hybridization could be inner octahedral , with the inert pairs popped into the central cavities . The underlying Bi(V) kernel is very strongly electrophilic and oxidizing [E0red = 1.759 V to Bi(III)] ; it has a powerful grip on the 6s2 inert pair . A low pressure synthesis might only produce the former non-metallic material ; very high pressure conditions would be required for synthesizing the desired superconducting form .

Bismuth(III) chloride and bismuth(III) bromide might also be examined in this context , but bismuth(III) fluoride , a relatively high melting ionic salt-like compound , would probably be unsuccessful . As with the iron(II) compound K4FeF6 mentioned above , the fluorides are too unreactive to transfer charge to the underlying Bi(V) kernel , and thus the hypothetical compressed BiV[**]F3 wouldn't be stable when it was cooled to room temperature and decompressed ; it would revert to ordinary BiIIIF3 [or maybe to a non-metallic ReO3 form with outer octahedral sp5 hybrid orbitals for the bismuth(III) , with localized inert pairs] :

In the above Table , the Bi(V) E0red Hal E0redvalue is a rough measure of the degree of electron transfer from the halide anion to the underlying Bi(V) kernel , and so is an approximate indicator of the strength of the BiHal coordinate covalent bonds . The Bi(V)–I coordinate covalent bonds are predicted to be the strongest , followed by the Bi(V)–Br and Bi(V)–Cl bonds . The negative value for the Bi(V)F bonds suggests that they will remain ionic in nature under all conditions . The relatively low melting points of BiCl3 , BiBr3 , and BiI3 are a good indication of covalent BiHal bonds in them . The dark brown-black color of the BiI3 , compared to the lighter color of the other Bi(III) halides , is a sign of substantial I> Bi charge transfer in it , with lesser charge transfer in the chloride and bromide , and none in the fluoride .

The same bonding situation occurs in lead(II) iodide , which has the hexagonal CdI2 layered structure :

This sketch was copied from the Wikipedia web page , Lead(II) iodide . Again , my thanks to the author of this sketch , and Wikipedia , for implied permission to reproduce it on this web page .

Lead(II) is isoelectronic with Bi(III) both are 6s2 so the valence electron distribution would be identical in normal PbI2 (outer octahedral) and in compressed Pb[**]I2 (inner octahedral) to that in the Bi analogues . When cooked under HPHT conditions the compressed product Pb[**]I2 would probably retain its CdI2 layered structure (with diminished lattice constants) , but now the electron pairs would be popped into the interlayer van der Waals space . These two dimensional channels , while making excellent pathways for the hopping electron pairs , would result in a more anisotropic electrical conductivity behaviour for PbIV[**]I2 than for the symmetrical BiV[**]I3 .

Several properties of the lead(II) halides are presented in the following table :

The lead(II) halides seem to be more ionic and less covalent in nature than the corresponding bismuth(III) halides . Very likely only lead(II) iodide would be satisfactory for compression in a HPHT experiment . Certainly PbF2 would be much too ionic , and PbCl2 and PbBr2 don't have a layered crystal structure like PbI2 .

The crystal structure of PbI2 is classified as hexagonal , and indeed the particles of lead(II) iodide when precipitated from an aqueous solution appear as sparkling , golden flakes having a hexagonal , snowflake-like appearance when viewed in magnification . See the nicely done YouTube video of “Golden Rain” [MP4 , 5524 KB , runtime 1:48] , providing an attractive classroom demonstration of the precipitation of an insoluble heavy metal compound from the metathesis reaction of a water-soluble lead compound and iodide (lead nitrate plus potassium iodide solutions) .

I should point out that an inner and outer tetrahedral electronic configuration , just like the inner and outer octahedral hybrid orbitals for K4Fe(CN)6 and K4FeF6 , and for BiI3 and BiV[**]I3 , are possible for the zinc blendes , as illustrated in the following sketch for Hg2+[**]Si2S4 :

If the synthesis of Hg2+[**]Si2S4 is attempted at ambient pressure , almost certainly the resulting product would be nonmetallic and electrically insulating , as shown in the B configuration above . If the structure has tetrahedral void spaces adjacent to the M0 atoms , the A configuration will be unlikely , as the 7s,p orbitals would be a much more confining interatomic void space . The C configuration shown above will be energetically favored if the 6s2 electron pairs can be popped into the adjacent void spaces .

The above analysis applies to the heavy metal cations with inert pairs such as Hg(0) , Tl(I) , Pb(II) , Sn(II) , and Bi(III) . For chemically reducing elements such as Zn0 and Li0 , there shouldn't be any innerouter tetrahedral coordination problem , because it's energetically favourable for these reducers to transfer their valence shell electron(s) into the tetrahedral voids . That leaves the corresponding Zn2+ and Li1+ cations , now tetrahedrally coordinated by sulfides or oxides , as the electronically inert Rare Gas kernels . More simply stated , Zn0 and Li0 don't have any inert pairs of electrons to remain stuck in their 4s and 2s orbitals , respectively .

The provision of suitable large , empty void spaces in the crystal structures of solid state materials is recommended as the guiding principle in the design of potential high temperature superconductors based on the formation of an electride under HPHT conditions . The ingenuity and creativity of experienced solid state chemists will be called upon in the future synthesis of such remarkable new materials .


Cu1+[*]Si2S4 and Li1+[*]Si2O4


Returning one final time to the zinc blendes , the synthesis and study of a representative zinc blende (or wurtzite) with zerovalent copper would be of great theoretical significance . If Cu1+[*]Si2S4 could be successfully synthesized , its electrical conductivity and superconductivity behaviour could conceivably reveal much about the inner workings of superconductivity .

Copper is a Noble Metal in the redox sense in that its Cu1+ and Cu2+ cations have positive standard reduction potentials , meaning that Cu0 is more thermodynamically stable than are its corresponding univalent or divalent cations . Cu(I) and Cu(II) are thus natural oxidizing agents like Hg2+ :

Cu1+ + e ----> Cu0 ; E0red = 0.521 V ;

Cu2+ + 2 e ----> Cu0 ; E0red = 0.3419 V ;

cf. Hg2+ + 2e ----> Hg0 ; E0red = 0.851 V .

Cu(I) forms charge transfer compounds and coordinate covalent S>Cu bonds with the sulfide anion , which is a natural reducer . Apparently Cu(II) oxidizes sulfide cleanly to disulfide and Cu(I) ; for example , CuS isn't Cu2+S2- but rather is a complex copper(I) disulfide .

Copper(0) can be inserted into the layered host lattice of titanium disulfide in a 1 : 2 molar equivalent ratio to obtain , after thermal treatment , the thiospinel Cu0Ti2S4 . The Cu0 atoms occupy the tetrahedral positions in the lattice ; the Ti(IV) atoms are octahedrally coordinated by the sulfides :

J. Padiou , D. Bideau , and J.P. Troadec , “Proprits Magntiques et lectriques de Thiospinelles Quaternaires” (“Magnetic and Electrical Properties of Quaternary Thiospinels”) , Chem. Abs. 92 , 225778j (1980) . Original article , which I haven’t been able to obtain : J. Solid State Chem. 31 (3) , pp. 401-405 (1980) [DOI] . A more recent research report on Cu0Ti2S4 by Professor Greedan's group at McMaster University , Hamilton , Ontario : N. Soheilnia et al. , “Crystal Structure and Physical Properties of a New CuTi2S4 Modification in Comparison to the Thiospinel”, Inorg. Chem. 43 (20) , pp. 6473-6478 (2004) [DOI] .

Cu0Ti2S4 is a covalentmetallic material ; its electrical conductivity can be attributed to the Cu(0) 4s1 valence electron , now probably in the 5s,p conduction band , as the S>Cu coordinate covalent bonds occupy the 4s,p orbitals . By analogy it might be possible to insert one equivalent of Cu(0) into two equivalents of the SiS2 host lattice in a HPHT synthesis :

Cu0 (m.p. 1085 C) + 2 SiS2 (m.p. 1090 C) -------- [HPHT] -------> Cu1+[*]Si2S4 .

Alternately , CuS or Cu2S could be used as the copper atom source , combined with elemental silicon and sulfur powders in the correct stoichiometric ratio :

CuS (m.p. 507 C , dec) + 2 Si0 (m.p. 1414 C) + 3 S0 (m.p. 115 C , b.p. 445 C)

-------- [HPHT] -------> Cu1+[*]Si2S4 ; or ,

Cu2S (m.p. 1129 C) + 2 Si0 + 3 S0 -------- [HPHT] -------> Cu1+[*]Si2S4 .

Since Si(IV) is certain to have a tetrahedral coordination by the sulfurs , Cu1+[*]Si2S4 should be a zinc blende or wurtzite , including the tetrahedral voids , and not a normal thiospinel . The copper atoms will be coordinated tetrahedrally by the sulfides as Cu(I) ; the S>Cu coordinate covalent bonds will occupy the coppers' 4s,p orbitals . The Cu(0) 4s1 valence electron will have the choice of entering the 5s,p conduction band or the tetrahedral voids adjacent to the copper atoms . Since the 5s,p orbitals will in essense be the spatially restricted interatomic voids around the Cu(I) kernels , the displaced electrons will probably enter the vastly larger tetrahedral voids where they will be energetically stabilized in the lattice .

Li0 could also be used in these singlet electron zinc blendes in place of the Cu0 . Since Li1+ is a low energy redox cation (unlike Cu1+, which is oxidizing) , it's unnecessary to use sulfide with it in the formulations ; oxide should be quite acceptable as the anion in the lithium compounds . Also , Li1+ has a strong preference for a tetrahedral coordination with oxides . Lithium is the most powerfully reducing element (3.04 V , Li0 > Li1+) , so it would be very energetically favourable for Li0 to pop its 2s1 valence electron into the tetrahedral voids . Of course , lithium metal itself would be too violent a reducer to use in the zinc blende synthesis ; as usual , the electrons that enter the voids would actually be provided by silicon powder as the reducing agent :

Li2O (m.p. 1437 C) + Si0 (m.p. 1414 C) + 1 SiO2 (m.p. 1722 C)

-------- [HPHT] -------> Li1+[*]Si2O4 .

Two possibilities emerge at this point . The first one is that Cu1+[*]Si2S4 would behave electrically much like the thiospinel Cu0Ti2S4 . If this was observed experimentally it would imply the copper 4s1 valence electrons were probably in the 5s,p conduction band .

The second possibility would be potentially momentous . Suppose the electrons in the voids ; they are totally free , unassociated electrons , perfectly mobile in the lattice under an applied potential difference . If these electrons weren't recaptured orbitally by the Cu(I) kernels at a higher temperature Cu1+[*]Si2S4 would be a fully functional ambient superconductor , but with singlet electron charge and energy carriers , not with Cooper pairs . That is , the free Drude electron gas , whether composed of electron pairs (as in Zn2+[**]Si2O4) or of singlet electrons (as in Li1+[*]Si2O4) , would be responsible for the phenomenon of superconductivity in all materials and under all physical and chemical conditions .

The possibility of Li1+[*]Si2O4 and Zn2+[*]Si2O4 being ambient superconductors brings to mind an interesting possibility : might Na1+[*]Si2O4 and Ca2+[*]Si2O4 also be superconducting ? Na1+ and Ca2+ typically prefer an octahedral , not tetrahedral coordination by oxides . Na1+[*]Si2O4 and Ca2+[*]Si2O4 might nevertheless be the starting point for research into the design and synthesis of superconducting soda-lime glass . Common soda-lime glass used for most glass products these days usually consists of approximately 74% SiO2 , 16% Na2O , 5% CaO , and several percent each of MgO , K2O , and Al2O3 . The zinc blendes are crystalline materials , of course , not amorphous glasses , but possibly the tetrahedral voids concept might be transferred to superconducting versions of sodium–calcium–silicon–oxides with voids . By analogy with fibre optic cable the invention of ambient superconducting glass filiaments that could be spun and braided into flexible “wires” would be of crucial importance and immense value to the energy technology of the twenty-first century .

Another potentially valuable application of the singlet electron zinc blendes might be in photovoltaic diodes (“solar cells”) , as shown in the following sketch :

Radiant energy such as sunlight can exert a finite , measureable pressure on material surfaces . This phenomenon has been the basis , for example , of the science fiction-like scheme for solar-powered vessels traveling in outer space between the planets , sailing like ships of the past on oceans , using the sunlight pressure on enormous light-gathering sails extending far out beyond the spaceship . Such sunlight pressure would push the singlet electrons near the surface of the superconducting wurtzite down deeper into the interior of the crystal , where they could pair up with other singlet electrons . The solar energy pressure would thus create a positive , electron-depleted zone on the “sunny side” of the material , and a negative electron-enhanced zone underneath . As long as the sunlight falls on the upper surface , a potential difference will be established between the two terminals attached to the diode . This p.d. can be tapped to draw off electrons from the lower negative terminal , moving them through the work circuit and then back again to the upper positive terminal . The photovoltaic cell would thus act like the water wheels in the mills of bygone eras . The river water flowing over the wheels turned them to provide mechanical energy for the mills . Similarly the sunlight's radiant energy would cause the singlet electrons in the illuminated diode to flow ceaselessly around and around in the work circuit as the diode's electrical energy is utilized .

If any of the zinc blendes with tetrahedral voids discussed above was discovered to be a fully functional ambient superconductor , not only the scientific but the technological implications would be profound . The real Superconductor Revolution will have been launched . Ironically it was also the crystallographically related silicon and germanium (and zincblendes such as gallium arsenide) that powered the Semiconductor Revolution beginning in the late 1940s and has had such an immmense impact on modern society .

A new picture of superconductivity has emerged from this web page . It hypothesizes that what really causes superconductivity isn't the Cooper pairs as such , but what the Cooper pairs are made of : Drude electron gas , which is comprised of free electrons unassociated with their parent atomic kernels . More specifically , superconductivity is caused by the flow of Drude electron gas through the interatomic voids of a solid . In contrast , conventional electrical conductivity is caused by Fermi-Dirac electrons , which can flow through the lattice in the crystal orbital , but which are firmly associated with their atomic kernels .

Solid state chemists can test this hypothesis by designing and synthesizing crystal containers for the Drude electron gas . Spacious voids , specifically channels of void space , must be incorporated into the lattice design to comfortably accomodate pairs of electrons that have been popped off metal atom kernels under high pressure and temperature conditions . Once in the void channels adjacent to the parent kernels the electron pairs will be free of them . They will then be the Drude electron gas , capable of pouring through the void channels in the lattice under an applied potential difference , flowing without resistance through the solid . There is no theoretical upper limit to the transition temperature of the superconductivity of the Drude electron gas . The material containing it will be superconducting as long as the lattice and the void channels remain intact and unblocked .

This is a simple , beautiful picture of a natural phemonenon whose explanation has defied researchers for somewhat over a century . As the poet John Keats once wrote when contemplating his Grecian urn ,

Beauty is truth , truth beauty that is all

Ye know on earth , and all ye need to know.

That's the poetic way of expressing Occam's Razor : The simplest among several different hypotheses is likely the correct one . I believe that Nature is fundamentally simple , and that humans make it seem extremely complex and incomprehensible in their clumsy efforts to understand it . In this same philosophical vein I also think that unless you can explain something in plain , simple , nonmathematical language you don't really understand it . And finally , here's a plain , simple truth : superconductors are chemical substances . The people who make them are chemists . In order to make new ones chemists have to understand superconductors in plain , simple , chemistry terms . That has always been my objective , as a chemist , for the past twenty-five years or so I've been studying the chemistry of metallic solids , including superconductors . I hope I have achieved that objective in this web page , too .


Related web pages in this series about Drude electron materials :

Perovskites Designed as Drude Metals and Ambient Superconductors” ;

Rocksalts Designed as Super-electrides , Drude Metals , and as Possible High Temperature Superconductors” ;

Chromium as the Guest Atom in Super-electride Drude Metals” ;

Lead , Tin , and Bismuth as the Guest Atoms in Super-electride Drude Metals” ;

* Betaines and Electrides : From Sugar Beets and Baby Shampoo to Superconductors ; and ,

* Drude Electron Materials Having Rutile and Layered Structures” .


[ Index Page ] [ Contact ]