Mixed-Valent Inert Pair Systems as Superconductor Candidate Materials


In two recent Chemexplore web pages (Inert Pairs and Thiospinels) various compounds of mercury(0) , thallium(I) , lead(II) , bismuth(III) , and tin(II) were proposed as possible new superconducting materials . Their crystal structures were chosen so that their ns2 inert pairs of valence shell electrons might be promoted up into the (n+1)s energy level . There they would be released from the coulombic grip of the electrophilic , oxidizing parent kernels , being dispersed as the electron gas of the metallic bond throughout the interatomic voids of the crystal lattice . It was hoped that the inert pairs of electrons would remain more or less associated in the (n+1)s,p conduction band (metallic bond) , in effect forming Cooper pairs as the charge and energy carriers . The new materials could thus be superconducting , possibly at fairly high temperatures , and maybe even at room temperature .

All of the compounds proposed in those web pages as metallic solids and superconductor candidates were homovalent ; that is , their electronically active atoms were in a single valence state . The following presentation will revisit the same heavy metal elements exhibiting the inert pair effect , but this time their mixed-valent analogues , specifically with (n) and (n+2) valence state components , will be studied . Various hypothetical compounds with Hg(0)/Hg(II) , Tl(I)/Tl(III) , Pb(II)/Pb(IV) , Bi(III)/Bi(V) , and Sn(II)/Sn(IV) will be examined as potential new metallic solids .

As discussed in the earlier web pages , the sulfide anion will coordinate to the metal cations so that the strongly oxidizing and electrophilic nature of the underlying ns0 kernels can be neutralized by the S2, which is a natural reducing agent . The “neutralized” kernels will no longer be attractive to the promoted inert pairs of electrons , which will remain in the metallic bond . In simple terms , the sulfide anions are required to stabilize the metallic state of the materials during and after their synthesis . The proposed new compounds are expected to be covalent–metallic ; that is , they will have underlying strong metal–sulfur covalent bonds forming their skeletal structures , and a much weaker metallic bond at a higher energy level permeating the interatomic voids in the lattice .

Very likely high pressure–high temperature (HP–HT) conditions , such as can be attained only in an anvil type of press , will be required to synthesize materials with such covalent–metallic bonds . A conventional high temperature synthesis at ambient pressure would likely result in ionic bonds in the products , with no promotion of the inert pairs up into the (n+1)s energy level . Very high pressure applied to the reactant atoms could compress them to the point where the sulfide anions surrounding and coordinating the heavy metal atoms are able to pop forcibly promote the inert pairs from the ns level up into the (n+1)s energy level , thereby creating the metallic bond in the crystal . The (n+1)s energy level in the heavy metals is the 7s level (6s for tin) , which for all intents and purposes is the interatomic void space in the lattice .


Hg , Zn , and Cd(0)–(II) Compounds


Monatomic monovalent mercury cations are unknown ; of course , dimeric monovalent mercury atoms , (Hg2)2+, are found in the common mercurous cations such as in calomel , Hg2Cl2 (Cl–Hg–Hg–Cl) . However , it might be possible to produce monatomic monovalent mercury cations , Hg1+, in a metallic solid , with the 6s1 valence electrons delocalized throughout the lattice in its 7s orbitals . The thiospinel HgAlTiS4 , written as Hg1+Al3+TiIVS4 to reveal the metal atom valence states , could provide a suitable “crystal container” for the metallic Hg1+ . Such mercury could also be formulated as the mixed-valent Hg00.5–Hg2+0.5 .

Mercury metal and the three component sulfides , b-HgS (the black form , having the zinc blende crystal structure) , Al2S3 , and TiS2 , would be combined together under HPHT conditions (anvil-type press) :

Hg0 (b.p. 357 C) + b-HgS (black , m.p. 850 C) + Al2S3 (m.p. 1100 C) + TiS2

-------- [HPHT] -------> HgAlTiS4 .

If the product HgAlTiS4 was a pure compound with the normal thiospinel structure , its mercury atoms would be tetrahedrally coordinated by the sulfide anions : HgtetAloctTioctS4 . As mentioned above , the reducing sulfides would coordinate the oxidizing and electrophilic mercury(II) kernels , and in doing so would promote the 6s2 valence electrons up into the 7s,p orbitals :

The question then becomes : to what extent will the 6s2 electrons – now 7s2 – remain associated as pairs in the metallic bond (7s,p conduction band) ? Will the Fermi-Dirac distribution “smear them out” over the physical dimensions of the lattice ? If that happened , HgAlTiS4 would behave much like a conventional metallic solid , with genuine Hg1+(now 7s1) . However , if the inert pair effect continued to cause their association in the metallic bond at higher temperatures (ie. with Hg00.5–Hg2+0.5 , now 7s2) , the material could be superconducting , possibly above the liquid nitrogen boiling point (77 K) , and maybe even at room temperature . That's why mercury is especially interesting , as its inert pair effect is remarkably strong at ambient conditions and above .

Similarly zinc and cadmium , which like mercury also have a pronounced M(0)–M(II) chemistry , might be used as the tetrahedrally coordinated metal atom in a thiospinel compound . Zinc(II) in particular strongly prefers a tetrahedral coordination by anions and other ligands . Both Zn2+ and Cd2+ are low energy redox species ; that is , they are neither oxidizing nor reducing in nature , unlike the strongly oxidizing Hg2+ . The analogous oxide spinels of Zn and Cd might therefore also be as stable as their corresponding sulfides , and could be investigated as novel metallic solids and as possible superconducting materials :

Zn0 (m.p. 420 C , b.p. 907 C) + ZnS (m.p. 1700 C) + Al2S3 (m.p. 1100 C) + TiS2

-------- [HPHT] -------> ZntetAloctTioctS4 , ie. Zn00.5Zn2+0.5AlTiS4 ;

Zn0 (m.p. 420 C , b.p. 907 C) + ZnO (m.p. 1974 C) + Al2O3 (m.p. 2054 C)

+ TiO2 (m.p. 1843 C) -------- [HPHT] -------> ZnAlTiO4 ;

Cd0 (m.p. 321 C , b.p. 767 C) + CdS (m.p. ~1480 C) + Al2S3 (m.p. 1100 C) + TiS2

-------- [HPHT] -------> CdAlTiS4 .

Cd0 (m.p. 321 C , b.p. 767 C) + CdO (m.p. 1559 C , subl.) + Al2O3 (m.p. 2054 C)

+ TiO2 (m.p. 1843 C) -------- [HPHT] -------> CdAlTiO4 .

Note that both zinc and cadmium form dimeric (M–M)2+ cations , like the mercurous cation [although they are less stable and more reactive than (Hg2)2+] . For example , (Zn2)2+ cations are produced by the solution of zinc metal in molten ZnCl2 , and similarly (Cd2)2+ cations have been observed :

D.H. Kerridge and S.A. Tariq , “The Solution of Zinc in Fused Zinc Chloride”, J. Chem. Soc. A 1967 , pp. 1122-1125 ; J.D. Corbett , “The Cadmium(I) Ion Cd22+. Raman Spectrum and Relationship to Hg22+ ”, Inorg. Chem. 1 (3) , pp. 700-704 (1962) .

However , it might be possible to synthesize genuine monoatomic , monovalent M1+ cations (M = Zn , Cd , and Hg) in nonmolecular , solid state structures under HP–HT conditions . In such infinite atomic lattices the M1+ cations are always separated by various types of anions , thereby preventing their dimerization to the covalent (M–M)2+ species .

The M3OX (M = Zn , Cd , and Hg ; X = S , Se , and Te) antiperovskites were briefly alluded to in the Iron web page . The M atoms are M2+–M0–M2+ , with the NIOS (non-integral oxidation state) valence of 1.33+ ; that is , their sigma XO metallic bonds (ns,p conduction bands) would be only 33% filled with free electrons . The antiperovskite crystal structure is a sort of “inside-out” perovskite structure , with a large anion (sulfide , selenide , telluride) occupying the central voids , and with the metal cations and smaller anions (oxides) in reversed positions in the lattice relative to those in the perovskites :

The 3D graphic was a “print screen” of the screen saver image on my computer's monitor . It more realistically represents a perovskite , in which the metal cations (now blue spheres) are small and the oxide anions (now green spheres) are medium-sized .

The oxygen atoms are octahedrally coordinated by the M atoms in this structure . Since oxygen can't form an octahedral hybrid orbital , the M–O bonds can't be covalent ; the oxygens in this hypothetical antiperovskite compound must therefore be oxide anions . The large central chalcogenides are also anions . The M atoms are thus M2+ cations , with their ns2 valence electrons promoted into the (n+1)s frontier orbitals , as the coordinating anions sterically block the ns and np orbitals , making them energetically inaccessible to the valence electrons . Alternately , the nucleophilic anions bond to the electrophilic M2+ kernels using the cations' ns and np orbitals , and displace the ns2 electrons up into their corresponding (n+1)s,p orbitals .

Synthesis of the M3OX antiperovskites might be attempted by the combination , under HPHT conditions , of the M elements and their oxides and chalcogenides :

Zn0 (m.p. 420 C , b.p. 907 C) + ZnO (m.p. 1974 C) + ZnS (m.p. 1700 C)

-------- [HPHT] -------> Zn3OS ;

Cd0 (m.p. 321 C , b.p. 767 C) + CdO (m.p. 1559 C , subl.) + CdS (m.p. ~1480 C)

-------- [HPHT] -------> Cd3OS ;

Hg0 (b.p. 357 C) + HgO (m.p. 500 C , dec.) + b-HgS (black , m.p. 850 C)

-------- [HPHT] -------> Hg3OS ; and similarly for the Se and Te analogues .

Tolerance factors can apply to both perovskites and antiperovskites . The following Table presents the tolerance factors calculated for the nine possible M3OX antiperovskites under consideration :

In calculating the tolerance factors listed in the above Table the crystal ionic radii of the large central chalcogenide anions were for six-coordination , as provided by the CRC Handbook of Chemistry and Physics , 87th edition , 2006 . However , in the M3OX antiperovskites they would be twelve-coordinated by the M2+ cations . Generally the ionic radius increases with increasing coordination , so the actual radii of the chalcogenide X anions would probably be larger than those used in the calculations . According to the Goldschmidt equation , an increase in the size of the central large atom , all other factors remaining unchanged , will increase the tolerance factor . Very likely , then , all nine of the hypothetical M3OX antiperovskites tabulated above would have a cubic crystal symmetry .

Zinc , cadmium , and mercury chalcogenides all have the cubic zinc blende crystal structure :

Above : the cubic zinc blende (sphalerite) structure , in which both the M and X atoms have a tetrahedral coordination . This image was copied from the Wikipedia web page , “Cubic Crystal System”. I thank the author of this sketch , and Wikipedia , for implied permission to reproduce it on this web page .

Mixed-valent zinc blende MX compounds with M0–M2+ (M = Zn , Cd , and Hg ; X = S , Se , and Te) might be synthesized by combining x M2+ and (1-x) M0 to make M(1+x)+ in situ , and adding sufficient fluoride anion to compensate for the reduced overall cation charge :

x MX + (1-x) M0 + (1-x) MF2 -------> M(1+x)+ [XxF1-x](1+x) ,

where x = a mole ratio taken experimentally between 0 and 1 .

For example ,

x ZnS (m.p. 1700 C) + (1-x) Zn0 (m.p. 420 C , b.p. 907 C) + (1-x) ZnF2 (m.p. 872 C)

-------- [HPHT , inert atmosphere] -------> Zn(1+x)+ [SxF1-x](1+x) , x = 0 to 1 .

That is , the substrate ZnS is being doped with increasing mole ratios “x” of the chemically reducing dopant “ZnF”, which is prepared in situ from Zn0 + ZnF2 . These electron doped zinc blende compounds were discussed in greater detail in the earlier Antifluorites web page .


Tl , In , and Ga(I)–(III) Compounds


Gallium , indium , and thallium all have known monovalent and trivalent species , but no stable divalent state . For example , the compound “GaCl2” has been reported , but on close examination has been shown to actually be GaCl–GaCl3 (a Robin-Day Class I mixed-valent compound) . Thallium(I) and thallium(III) are both very stable , well-known valence states of the element , with a variety of compounds for both atomic configurations . Thallium(I) compounds are ionic , and somewhat resemble those of potassium , to the extent where Tl1+ can readily substitute for K1+ in the body . Tl , as a heavy metal , can strongly bond to the sulfur atoms in proteins and enzymes and thereby deactivate them , and that makes thallium compounds highly toxic to most organisms [thallium(I) sulfate , resembling potassium sulfate , has been used in the past as a rodenticide] . It also makes thallium compounds quite unpopular with chemists , who generally shy away from them !

Thallium(III) compounds have mostly covalent properties . Tl(III) is a powerful oxidizer , with E0red = 1.252 V to Tl(I) ; thallium(III) nitrate , “TTN”, is used as an oxidizing agent in organic chemistry . The mixed-valent Tl(I)0.5 + Tl(III)0.5 = Tl(II) might be examined as both the tetrahedral and octahedral components of a normal spinel . In the former case the duo could be combined with two octahedral components such as La3+ :

Tl(I) + Tl(III) + [S] + La2S3 ---------> (Tl1+0.5 Tl3+0.5)tet (La3+ La3+)oct (S4)8.

Note : Lanthanum sulfide , La2S3 , while commercially available (Alfa-Aesar) , is rather expensive . However , the crystal ionic radius of La3+ (r = 1.03 ) is the largest of all the trivalent cations , so it will be assured of filling the octahedral voids in the spinel structure . The smaller Ga3+ (0.62 ) , In3+ (0.80 ) , and Tl3+ (0.89 ) cations will then occupy the tetrahedral voids in it , as required for these compounds . Values of the ionic radii cited were per Shannon and Prewitt , and were for six (octahedral) coordination .

Both thallium(I) and thallium(III) sulfides are known compounds ; the former is a blue-black crystalline material , m.p. 457 C , whose water solubility is 0.2 g/L at room temperature . The latter compound , Tl2S3 , is rather obscure . It has been described as a black , amorphous solid , m.p. 200 C (in N2) , and is also water-insoluble . The two sulfides might be co-precipitated from a water solution of equimolar quantities of thallium(I) and (III) nitrate by the addition of (or addition to) a sulfide solution :

TlNO3 (0.5 eq , aq) + Tl(NO3)3 . 3 H2O (0.5 eq , aq)

---------- [add Na2S . 9 H2O (aq)] ---------> [ Tl2S Tl2S3 = “TlS”] (ppt) .

The insoluble “TlS” precipitate would be washed , filtered , dried , and then combined with one equivalent of lanthanum sulfide in HPHT conditions to form the metallic thiospinel :

“TlS” + La2S3 (m.p. 2110 C) -------- [HPHT] -------> TlLa2S4 .

The Tl(I) component of TlLa2S4 would be coordinated by the surrounding sulfide ligands as the Tl(III) kernel , with the 6s2 inert pair promoted up into the 7s orbitals . The chemically reducing sulfides would donate charge to the electrophilic , oxidizing Tl(III) kernels , thus neutralizing their oxidizing power and stabilizing the metallic bond (7s,p conduction band) in the spinel lattice .

If “Tl(II)” was used as the octahedral component of the spinel , a suitable tetrahedral companion would have to be found for it . As discussed in the Thiospinels web page , the silicon(IV) in silicon disulfide might satisfy this requirement :

SiS2 (m.p. 1090 C) + 2 “TlS” -------- [HPHT] -------> SiTl2S4 .

TlS itself , when subjected to HPHT conditions , might adopt the zinc blendewurtzite crystal structure . The reducing sulfide anions , with eight valence electrons each (3s2 3p6) , could form tetrahedral sp3 hybrid orbitals ; the oxidizing Tl(III) kernels (6s0 6p0) could form similar sp3 hybrid orbitals , which are empty and can receive charge from the sulfides . The three dimensional S>Tl coordinate covalent bond network would form the zinc blende or wurtzite lattice . Meanwhile , the 6s2 inert pairs on the Tl(I) components of TlS would be promoted up into the 7s orbitals , thus forming the metallic bond in the solid .

TlS could resemble the covalent–metallic compound tin(III) phosphide , SnP , which is an excellent electrical conductor at all temperatures , and whose cubic rocksalt form becomes superconducting close to Absolute Zero (Tc ~ 4 K) . The free electrons in the metallic bond of SnP behave as singlet electrons in the normal state . Even though Sn(II) demonstrates an inert pair effect , there is no apparent inert pair coupling at higher temperatures in SnP . Very likely a similar situation would be observed in TlS , if indeed it proved to be metallic after a HPHT synthesis .

The univalent state is least stable in gallium , and somewhat more so in indium . Mixed-valent M(I)–(III) gallium and indium compounds analogous to the thallium ones outlined above could be investigated as novel metallic , and possibly superconducting materials . Their syntheses might be accomplished by reproportionating the zerovalent elements with their corresponding trivalent reagents , and by direct combination of the elements concerned under HPHT conditions ; for example ,

1/3 Ga0 (m.p. 30 C , b.p. 2204 C) + 1/3 Ga2S3 (m.p. 1090 C) + La2S3 (m.p. 2110 C)

-------- [HPHT] -------> GaLa2S4 ;

1/3 In0 (m.p. 157 C , b.p. 2072 C) + 1/3 In2S3 (m.p. 1050 C) + La2S3 (m.p. 2110 C)

-------- [HPHT] -------> InLa2S4 ;

SiS2 (m.p. 1090 C) + 2/3 Ga0 + 2/3 Ga2S3 -------- [HPHT] -------> SiGa2S4 ;

SiS2 + 2/3 In0 + 2/3 In2S3 -------- [HPHT] -------> SiIn2S4 ;

Ga0 (m.p. 30 C) + 2 La0 (m.p. 920 C) -------> GaLa2 (alloy) ------->

[+ 4 S0 (m.p. 115 C , b.p. 445 C)] -------> GaLa2S4-------- [HPHT] -------> GaLa2S4 .

In0 (m.p. 157 C) + 2 La0 (m.p. 920 C) -------> InLa2 (alloy) ------->

[+ 4 S0 ] -------> InLa2S4-------- [HPHT] -------> InLa2S4 .

The combination of flowers of sulfur with metal powders often results in a violently exothermic reaction , so it might be prudent to first form an alloy of gallium and lanthanum (or of In and La) which presumably would be less reactive and more controllable when reacted with sulfur .

The two uncharacterized compounds “GaLa2S4” and “InLa2S4” would likely be non-metallic , and would require a HPHT treatment in an anvil type of press to convert them into the desired metallic spinels .

1/3 Ga0 + 1/3 Ga2S3 -------- [HPHT] -------> GaS (metallic zinc blende or wurtzite) ;

1/3 In0 + 1/3 In2S3 -------- [HPHT] -------> InS (metallic zinc blende or wurtzite) .

GaS and InS (from HPHT syntheses) are predicted to resemble the covalentmetallic SnP in their electronic properties . In their ordinary forms they (and TlS) are known , if somewhat obscure compounds . GaS is a yellow , crystalline solid , m.p. 965 C , and is a commercially available reagent (Alfa-Aesar , American Elements) . It clearly is nonmetallic , which isn't surprising when its crystal structure is examined :

This image was copied (with slight annotations) from the Wikipedia web page , “Gallium(II) sulfide”. I thank the author of this sketch , and Wikipedia , for implied permission to reproduce it on this web page .

GaS is a layered material with the GaS sheets bonded together into double layers by GaGa covalent bonds . Bulky , stereochemical lone pairs of electrons (not shown in the sketch) on the peaks of the sulfur pyramids protrude into the interlayer spaces . The chemical bonding in GaS is quite simple and straightforward if one assumes the transfer of a sulfur valence electron to its neighbouring gallium atom . Then the gallium has the four electrons required for the three GaS covalent bonds to its sulfur neighbours , and for the GaGa covalent bond :

If this yellow , nonmetallic , layered form of GaS was subjected to HPHT conditions in an anvil type of press it might be converted to a black , metallic form having an undistorted zinc blende or wurtzite crystal structure . One of the three gallium valence electrons will have been squeezed up into the Ga 5s,p orbitals , which can form the 5s,p conduction band (metallic bond) in the lattice :

This metallic GaS would be isoelectronic with ZnS , with the addition of the extra gallium valence electron , which must be accomodated in the empty 5s,p frontier orbitals above the lower energy GaS covalent bonds . Very likely nonmetallic InS and TlS could similarly be converted into analogous metallic zinc blende or wurtzite structures under HPHT conditions .

Since Ga(III) and In(III) are both low energy redox species , neither oxidizing nor reducing , their metallic oxide analogues to the above sulfides might also be stable after their HPHT synthesis . The compounds GaLa2O4 , InLa2O4 , TlLa2O4 , SiGa2O4 , SiIn2O4 , SiTl2O4 , GaO , InO , and TlO might thus be of some interest as novel metallic solids .


Pb(II)–(IV) and Sn(II)–(IV) Compounds


The thiospinel with octahedral Pb(II)(IV) , ZnIItet(PbIIPbIV)octS4 , comes to mind (Roman numerals indicate covalently-bonded atoms) . An additional equivalent of sulfur could be combined with the common PbS to make the uncommon PbS2 in situ :

ZnS (m.p. 1700 C) + 2 PbS (m.p. 1113 C) + S0 (m.p. 115 C , b.p. 445 C)

-------- [HPHT] -------> Zn(PbIII)2S4 .

PbIII in the product's formula is a formalism , as the Pb(IV) kernels , which are strongly oxidizing and electrophilic , would be octahedrally coordinated by sulfides , which neutralize their oxidizing power . The Pb(II) components' 6s2 inert pairs would be promoted up into the 7s orbitals , which would form the 7s,p conduction band in the lattice .

The small zinc cation strongly favours a tetrahedral coordination (as in the zinc blende ZnS) , and is probably optimal for filling the tetrahedral positions in the spinel lattice . Lead atoms , on the other hand , are undoubtedly too large to occupy tetrahedral voids in spinels ; they are suitable only for the octahedral positions .

The 5s2 inert pair in tin(II) might be promoted up into the 6s orbitals in ZnIItet(SnIISnIV)octS4 :

ZnS (m.p. 1700 C) + SnS (m.p. 881 C , b.p. 1210 C) + SnS2 (m.p. 600 C , dec.)

-------- [HPHT] -------> Zn(SnIII)2S4 .

Again , Sn(III) is really Sn(IV) + e , and Zn(SnIII)2S4 is predicted to be be a covalentmetallic compound . The reagent SnS2 is an interesting layered , graphite-like material ; it has a golden color and a lustrous , metallic appearance , but is an electrical insulator . It has been referred to in the past as “Mosaic Gold”, having been used as a golden paint pigment by medieval artists . SnS2 has the cadmium iodide crystal structure , with octahedrally coordinated tin(IV) atoms .

Zn(SnIII)2S4 would almost certainly be a metallic solid , and might also be superconducting . It could resemble SnP (Tc ~ 4 K) , mentioned above in connection with TlS , Ins , and GaS . Zn(SnIII)2S4 would be a most interesting material to synthesize and study .


Bi(III)–(V) in BiS2


Use of the Bi(III)Bi(V) mixed-valent couple in thiospinels seems impractical ; however , the hypothetical compound BiS2 , with fully reproportionated Bi(III)Bi(V) , should be metallic and possibly superconducting if the Bi(III) 6s2 inert pairs can be promoted up into the 7s orbitals by sulfide coordination . Such coordination could be either tetrahedral or octahedral upon promotion of the inert pairs . In the tetrahedral case the BiS2 could have the silica crystal structure , or one closely related to it , such as was discovered (by Prewiit and Young , 1965) for SiS2 compressed to 6065 Kbars (~ 6065,000 atmospheres) :

More likely , though , the Bi(IV) in BiS2 would be octahedrally coordinated by the sulfides . If the sulfides have a trigonal planar coordination by the bismuth atoms the material would have the rutile crystal structure (after the mineral rutile , TiO2) :

For rutile BiS2 , the blue spheres are octahedral Bi(IV) and the green spheres represent the trigonal planar sulfides (or sulfur atoms , in a covalently bonded lattice) .

BiS2 might alternately have the cadmium iodide crystal structure (as with TiS2 and SnS2 , for example ; GIF image , 24 KB) , in which the Bi(IV) atoms are octahedrally coordinated and the sulfur atoms have a trigonal pyramid coordination by the Bi atoms . However , the more compact rutile structure would probably be favoured under HPHT conditions rather than the lower dimensional layered CdI2 form .

BiS2 might be synthesized under HPHT conditions (anvil type of press) by the direct combination of bismuth and sulfur in a 1 : 2 gram-atom ratio , or possibly by the insertion of sulfur atoms into the lattice of bismuth(III) sulfide :

Bi0 (m.p. 271 C) + 2 S0 (m.p. 115 C , b.p. 445 C) -------- [HPHT] -------> BiS2 ;

Bi2S3 (m.p. 850 C) + S0 -------- [HPHT] -------> BiS2 .

The former method is probably the simpler and more economical approach to BiS2 .


SP2Tl4 , A Metallic Anti-thiospinel


Consider an equimolar combination of the two reagents , zinc oxide and zinc phosphide :

ZnO (m.p. 1974 C) + Zn3P2 (m.p. 420 C , 1100 C)

-------- [grind together , press pellet , heat] -------> OP2Zn4 .

Zinc phosphide is insoluble in water and alcohol , but is soluble in benzene and carbon disulfide . This unusual compound seems to have covalent , not ionic Zn–P bonds (both the zinc and phosphorus atoms are tetrahedrally coordinated) . It somewhat resembles anhydrous zinc chloride (m.p. 283 or 290 C) , which has a substantial covalent character . Zinc phosphide is used commercially as a rodenticide . It is also being studied as a substrate for semiconducting and photoelectronic devices , such as the photovoltaic sensor in solar cells (PDF , 509 KB) .

The lower melting phosphide would act as a flux to dissolve the more refractory ZnO . As the semi-molten mixture is cooled , it would crystallize into the hypothetical compound OP2Zn4 . In this material the positions of the metal and non-metal atoms are reversed relative to spinels : the smaller oxide anions would be tetrahedrally coordinated , the larger phosphide anions would be octahedrally coordinated , and the small zinc(II) cations would be tetrahedrally coordinated . If this thought experiment was carried out in practice and proved to be correct , OP2Zn4 could be said to have an anti-spinel crystal structure , the converse of the normal spinel structure .

Mixed-valent Tl1+Tl3+ might be incorporated into such an anti-spinel lattice as the metal atom components of the hypothetical compound SP2Tl4 : (S2)(P3)2 (Tl1+Tl3+ Tl1+Tl3+) . As a metallic solid with inert pairs (e22) promoted into the Tl 7s empty frontier orbitals its formula could be written as [SP2(TlIII)4–2(e22)] . The chemically reducing sulfide and phosphide anions would donate charge to the strongly oxidizing Tl(III) atoms , thus cancelling their electrophilic nature and stabilizing the metallic state of the material .

The anti-thiospinel SP2Tl4 , which would have considerable covalent bonding character , might be best synthesized by a direct combination of the three component elements under HPHT conditions :

S0 (m.p. 115 C , b.p. 445 C) + 2 P0 (red , m.p. 579 C , subl. 431 C)

+ 4 Tl0 (m.p. 304 C , b.p. 1473 C) -------- [HPHT] -------> SP2Tl4 .

Such a material would be quite interesting to study , both as a new type of crystal structure and as a metallic solid and possible superconductor .



[ Index Page ] [ Contact ]