Converting p-Block Element Inert Pairs into Cooper Pairs in High Temperature Superconductors



In an earlier Chemexplore web page (mixed-valent) and in two more recent ones (CuNiO and fluorides) the concept of inserting Noble Metals into unreactive host structures to obtain interesting new compounds with zerovalent metal atoms was introduced and explored to some extent . These novel materials are expected to be metallic solids , and could in some cases be superconductors at low temperatures . In this web page the theme is extended to a variety of post-Transition metal p-block elements in whose chemistry the inert pair effect is prominent (the references are presented at the end of the text , below . Underlined blue hyperlinks can be clicked when online to retrieve the article cited . The requested document will open in a new window) .

A novel enabling mechanism for high temperature superconductivity (HTS) is proposed in this report , together with the practical chemistry by which it might be implemented in readily accessible compounds . Two other enabling mechanisms for superconductivity include the classic BCS phonon-assisted Cooper pair formation near Absolute Zero , and AFM (antiferromagnetic) induction (GIF image , 77 KB) for facilitating Cooper pair condensation at relatively high temperatures (above the boiling point of liquid nitrogen at 77 K) .

In this third enabling mechanism the inert pairs of ns2 valence shell electrons in certain p-block elements are forced , by a combination of chemical pressure (from surrounding nucleophilic anion ligands) and human-supplied mechanical pressure (anvil press) into empty frontier (n+1)s0 orbitals . In these outer orbitals they can escape the bonding attraction of the underlying electrophilic kernels . Now above EF , the Fermi level , they become mobile , itinerant Cooper pairs in the superconducting state . Hopefully the pairing strength the “two-ness” of the inert pairs below EF will be retained to a certain extent in the Cooper pairs above EF , resulting in exceptionally high transition temperatures , Tc , for the new compounds .

Mercury displays a strong inert pair effect in its chemistry , so its zerovalent atom insertion compounds would be very interesting to examine in the present context . While each of the elements is quite unique and fascinating in its own right , mercury is particularly so for me . Who , after all , isn't delighted and enchanted by globules , puddles , and streams of a silvery , shiny , liquid metal ? This highly unusual physical property extraordinarily low melting and boiling points (39 and 357 C , respectively) for a metal – is undoubtedly the consequence of the ns2 inert pair effect observed with many post-Transition metal p-block elements :

Because mercury's valence shell electrons are exclusively 6s2 , with no 6p valence electrons and no participation of its 5d10 shell electrons (which are extremely inert , electronically) , the 6s2 electrons are its inert pair . That makes the mercury atom chemically rather unreactive . The corresponding Hg2+ cation is a moderately strong oxidizer , so mercury can be classed as a Noble Metal , at least in the redox sense .

The extraordinary stability of the 6s2 inert pair in mercury is illustrated by the lack of any recognized monoatomic mercury(I) species (that is , with 6s1) . All known Hg(I) compounds have diatomic (Hg2)2+ molecular cations (for example , as in calomel , Hg2Cl2 , which is Cl–Hg–Hg–Cl) . The mercury atoms in these “Siamese twin” mercury(I) cations have completed 6s2 valence shells .

The 6s2 inert pair effect in zerovalent mercury atoms results in a very feeble sort of metallic bond in the liquid . In all the elementary metals having a ns2 valence shell , such as those in the IIA/2 and IIB/12 families (Alkaline Earth metals and zinc group , respectively) , some leakage of the ns2 shell electrons into the energetically accessible np0 orbitals always occurs , thus forming the s-p conduction band , which is the metallic bond in these elements . Without such an electron leakage these divalent elements would be insulators or semiconductors , as their sigma XO (crystal orbital = “polymerized molecular orbital” = conduction band = metallic bond) would be completely filled with electrons . However , mercury's pronounced inert pair effect largely negates this ns2 to np0 electron leakage , resulting in its anomalously low electrical conductivity compared to the lighter members of its family , zinc and cadmium :

zinc : 166,389 ohm-1-cm-1 (298 K)

cadmium : 147,059 ohm-1-cm-1 (273 K)

mercury : 10,406 ohm-1-cm-1 (298 K)

Data were from the CRC Handbook of Chemistry and Physics , 87th edition , “Electrical Resistivity of Pure Metals”, pp. 12-39 to 12-40 (2006) .

With no 6s2 to 6p0 electron leakage at all mercury would be a colorless , very dense , nonmetallic gas , a sort of giant , bloated analogue of helium (1s2) . Liquid mercury has an appreciable vapor pressure even at room temperature , and is surprisingly volatile , forming exactly such a colorless , dense gas . The toxic nature of mercury necessitates its careful storage and handling , with an efficient ventilation (fume hood) . As it is , the element is barely able to function as a metal . Mercury's highly anomalous physical and chemical properties can be directly attributed to the strong inert pair effect occurring in its zerovalent atoms .

Frozen solid in a bath of liquid helium , mercury becomes superconducting at around 4 K . The only electrons that can become Cooper pairs in superconducting mercury are its 6s2 inert pairs , since it has no 6p electrons , and its 5d10 electrons are completely submerged inside the atomic kernel . As long as there is any 6s2 to 6p0 electron leakage Pauli paramagnetism will inhibit the formation of the Cooper pairs in the frozen mercury . According to the classical BCS theory , phonons in metallic solids assist in the formation of Cooper pairs in them at temperatures close to Absolute Zero . Such phonons may be responsible for finally halting the 6s2 to 6p0 electron leakage near 4 K , extinguishing the Pauli paramagnetism in the mercury , and permitting the formation of the Cooper pairs . The 6s2 inert pairs of electrons , normally below EF in mercury , are now all above it as Cooper pairs , and can become mobile and flow through the mercury crystal lattice under an applied potential difference across it .

There may also be a redox effect contribution to the retaining of the inert pairs below EF . In our chemical experience we know that as they are heated up , oxidizers become more strongly oxidizing , and reducers become more strongly reducing . That is , their reduction potentials Ered and oxidation potentials Eox , respectively , must be steadily increasing from their STP values , E0red and E0ox , as their temperature rises . The converse must also be true : as an oxidizer is cooled , its Ered must similarly decrease , eventually reaching zero at a certain low temperature .

The decline in electrode potentials with falling temperature is at times memorably demonstrated to motorists living in northern countries (like my own Quebec , Canada) on particularly frigid days in mid-Winter . Starting one's car – getting the engine turned over on ignition – can sometimes be quite difficult in such cold temperatures . The redox potentials of the Pb0 / PbO2 // H2SO4 in the car battery have fallen with the bitter cold , with a resulting low output voltage across its terminals . The battery sometimes can't provide enough power to crank the starter motor for the engine , which then needs to be externally boosted by power from another vehicle .

Hg2+ cation is a moderately strong oxidizer , with E0red = 0.851 V at STP (versus the SHE) . The zerovalent mercury atom consists of the Hg2+ cation spherically surrounded by the 6s2 inert pair of electrons . The strongly electrophilic Hg2+ kernel is exerting a powerful bonding attraction for the electronegative inert pair at ambient temperatures and above . As the mercury atoms are cooled , their Hg2+ Ered must gradually decline until at some point it becomes zero . Suppose the Hg2+ Ered = 0 at T ~ 4 K . Then the Hg2+ kernel ceases being electrophilic and loses its bonding attraction for the inert pair . At that point , the inert pairs rise above EF and become Cooper pairs , and the frozen mercury becomes superconducting .

Of course , the electrostatic (coulombic) attraction between the Hg2+ kernel and the inert pair electrons never changes over any temperature range . Thus , the Cooper pairs , while mobile and itinerant , don't stray very far from their respective kernels , just as in ordinary metallurgical metals the mobile , free electrons in their metallic bonds remain closely associated with their kernels as well . However , as soon as a potential difference (voltage) is applied across the ends of the metal or superconductor , its electronic charge carriers (electrons and Cooper pairs , respectively) start “hopping en mass” downfield from kernel to kernel throughout the lattice . The “picture analogy” of electrical conduction I like is that of a pinball machine , with the metal balls bouncing off the posts as they roll down the slope of the gameboard . In all normal circumstances the electrical charge carriers remain closely associated , in the coulombic sense , with their parent kernels .

The objective of the new compounds proposed in this report , and of the inert pair enabling mechanism for HTS , is to duplicate the superconductivity in frozen mercury at 4 K , but at much higher temperatures , and possibly even at ambient temperature . The electrophilic bonding attraction of the Hg2+ kernel to the inert pairs will be cancelled by the clustering about Hg2+ by a number (usually four , tetrahedral) of nucleophilic anion ligands ; these anions (commonly sulfide , generally chalcogenide) are natural reducers , and will transfer charge to the Hg2+ , thus nullifying most of its electrophilic nature . The anions will also repel the promoted inert pairs away from the kernels by their strongly electronegative fields , and shield them from the kernels' positive charge .

While purely speculative , the hope is that the inert pairs will be converted from the localized 6s2 valence electrons below EF to the delocalized , “free” Cooper pairs above EF in the outer frontier orbitals to which they have been promoted . Their promotion should be enabled by the nucleophilic anions pressing down on the Hg2+ kernel – “chemical pressure” – and by a “mechanical pressure” of the compression of the crystal lattice in a powerful anvil type of press .

I mentioned a second picture of the promoted inert pairs in the Fluorides web page . In that alternate scenario they become the electron gas flowing in the interatomic voids in the lattice . It could be that as Cooper pairs the inert pairs are no longer in any sort of orbitals ; only singlet electrons are , in the orbitals forming the metallic bond XO , where they are subject to the Fermi-Dirac distribution (this may be true of superconductors in general) . Because the Cooper pairs aren't in orbitals (so technically they aren't in a metallic bond , either) they can avoid being “smeared out” across the lattice by the Fermi-Dirac distribution .

That's the chemistry picture” of the zerovalent mercury compounds as HTS candidate materials . Can they be synthesized in actual laboratory practice ? As pointed out in the CuNiO web page , the stability and “practicality” of any such zerovalent atom insertion compounds will depend on the coordinations of their electropositive element atoms (the metal ones) with the electronegative element components (the non-metal ones) . The M–X coordinations must be “natural” and not “forced”. In a covalently-bonded system the coordinations must generally adhere to those described by the Valence Bond theory , eg. sp , linear ; sp2 , trigonal planar ; sp3 , tetrahedral , etc. There is probably a greater latitude for diverse coordinations in ionically bonded systems , given the omnidirectional electrostatic field of the common anion ligands .

The “natural” coordinations for each element in its various valence states are a function of the three-dimensional geometry of that element's valence shell orbitals . Also , in covalently-bonded systems there is the additional consideration of the hybridization of two or more native orbitals to form the “directional hybrid orbitals” (GIF image , 42 KB) described by the Valence Bond theory . Such hybrid orbitals will set the bond angle geometry for the atoms in such covalent systems . From their accumulated chemical knowledge chemists are quite familar with at least the more common coordinations of the elements in their usual valence states . In Hg–oxygen systems the O–Hg–O bond is universally linear , while in Hg–halogen and Hg–chalcogen systems occasionally linear but mostly tetrahedral Hg–X coordinations are observed .

Mercury(II) oxide , HgO , would be unsuitable as a precursor for any novel candidate insertion compound . It decomposes at ~ 500 C , but in reality it is a powerful oxidizer even at room temperature . Apparently a variety of metals (and other substances) react exothermically with HgO with considerable violence :

“[HgO] Reacts violently with reducing agents , chlorine , hydrogen peroxide , magnesium (when heated) , disulfur dichloride , and hydrogen sulfide . Mixtures with metals and elements such as sulfur and phosphorus are shock-sensitive” – from this Inchem web page about HgO .

Mercury(II) oxide has also been used as a dehydrogenating oxidizer in certain organic chemistry reactions at STP .

Mercury(II) chalcogenides might be safely tried as precursors for possible new zerovalent Hg compounds . While the oxide anion is essentially a low energy redox species (neither oxidizing nor reducing in nature) , the chalcogenide anions are natural reducing agents , that when combined with Hg2+ form charge-transfer compounds rather than the respective elements :

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

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

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

The rather large , positive cell potential E0T for the Hg2+ / S2- charge transfer indicates that the net redox reaction as written would be thermodynamically spontaneous at STP .

The standard redox potentials cited are versus the SHE at STP , and were from : D.R. Lide (ed.) , CRC Handbook of Chemistry and Physics , 87th edition , CRC Press / Taylor & Francis , Boca Raton (FL) , 2006 ; P. Vansek (ed.) , “Electrochemical Series”, pp. 8-20 to 8-25 . Another useful reference with redox potentials is : A.J. Bard (ed.) , Encyclopedia of Electrochemistry of the Elements , various volumes , Marcel Dekker , New York , ca. 1973-1986 .The Wikipedia web page “Table of Standard Electrode Potentials” is a helpful online reference in this regard . For a convenient tabulation of oxidizing metal oxides and their E0red values , see this GIF image (45 KB) .

Thus , in HgO the oxygen atoms have no reducing nature and the full fury of the Hg(II) oxidizing power all 0.851 V of it , quite substantial will fall on any external reducing agent it comes in contact with . In the chalcogen compounds like HgS this oxidizing power has been effectively cancelled or muted by the reducing nature of the chalcogen anions . It should therefore be possible to react Hgchalcogen compounds with mildly reducing reagents in a safe , effective (ie. non-explosive) manner .

Having selected suitable precursor Hg(II) reagents , the next task is to design acceptable container” crystal systems for the Hg(0) . Such containers must be compatible with the preferred geometry of the base Hg2+ cations' coordination by the chalcogenide atoms , which usually is sp3 tetrahedral . The objective will then be to surround the Hg0 with four chalcogen (eg. sulfur) atoms in a tetrahedral coordination , and forcibly “pop[promote] the 6s2 inert pair into the lowest energy vacant frontier orbital (LUMO) available . Since a tetrahedral coordination involves the sp3 orbitals on the mercury atoms (6s + 6px,y,z) , the most probable frontier orbital for the “popped” 6s2 inert pair would be mercury's 7s native atomic orbital .


Zerovalent Mercury in Thiospinels and Wurtzite Sulfides


The metallic solid CuTi2S4 , a normal thiospinel , was prepared by French researchers by inserting copper(0) atoms into the layered host lattice of titanium(IV) sulfide in a 1 Cu0 : 2 TiS2 ratio . The intercalated material Cu0.5TiS2 was heated to convert it into the spinel . The zerovalent copper atoms occupied the tetrahedral voids in the structure , while the titanium(IV) atoms were octahedrally coordinated by the sulfide anions .

McKelvey and co-workers described the intercalation of mercury(0) in titanium disulfide in a series of research reports during the 1990s . The mercury atoms readily soaked into” the interlayer voids between the TiS2 sheets . Simply by exposing the TiS2 host to the mercury vapor in a closed tube the researchers were able to prepare adducts at various mercury doping levels , from x = 0 to ~ 1.24 in Hg0xTiS2 :

“Triply distilled Hg (< 5 ppm foreign metals) and TiS2 were loaded into quartz ampules in a drybox , evacuated to < 10-4 Torr at –196 C , and flame sealed . The gradual uptake of Hg by the host is apparent at ambient temperature . Samples were homogenized by annealing for 2 days at 320 C , followed by slow cooling to ambient temperature” (p. 14 of the 1992 Chem. Mater. article) .

The mercury atoms were only weakly bonded to the sulfur atoms of the TiS2 host :

“TGA of HgxTiS2 shows that Hg begins to slowly deintercalate at about 170 C , and substantial deintercalation occurs only above 250 C and is complete at 330 C .....” (p. 14) .

From various considerations McKelvey concluded ,

“The above observations indicate there is essentially no or very little charge transfer associated with the intercalation of Hg into TiS2 . A small amount of covalent electron exchange is also possible based on these studies . Hence , we conclude that the primary driving force for Hg–TiS2 intercalation is not the redox reaction of individual guest species with the host , as has been generally assumed for metal–TMD intercalation” (p. 15 ; TMD = Transition metal dichalcogenide) .

Given the Noble Metal nature of mercury this conclusion isn't too surprising . A Valence Bond analysis of the electronic structure of the TiS2 host might shed some light on the Hg–S bonding in HgxTiS2 :

Titanium disulfide (discussed in the Layered Chemexplore web page) is a metallic solid with a lamellar morphology like graphite . It has a golden-yellow colour with a bright metallic luster , and is a fair electrical conductor , with an ambient conductivity of about 1400 ohm-1-cm-1 . TiS2 is a True Metal [GIF image , 41 KB] , with an inverse electrical conductivity–temperature relationship from room temperature down to near Absolute Zero . It has the cadmium iodide crystal structure with S–Ti–S layers :

Very likely the titanium atoms' 3d2 4s2 valence electrons have been pushed into the empty 4p frontier orbitals by the coordinating sulfur atoms' electron pairs . This Valence Bond analysis suggests that the sulfur atoms' 3s orbitals are empty , because of their trigonal pyramid (p3) hybrid orbitals , whose electron pairs – all six of sulfur's 3s2 3p4 valence electrons – are fully engaged with the titaniums .

A well-known chemical property of the Transition metal dichalcogenides , including TiS2 , is that they will readily accept and intercalate electron donors (reducers) , but won't form any such intercalation compounds with electron acceptors (oxidizers) [unlike graphite , which will accept certain ones of both varieties] . This is because , on the one hand reducers can donate some of their electrons into the empty 3s orbitals on the sulfur atoms , while on the other hand the “empty” sulfurs have no electrons to donate to oxidizers .

In the case of McKelvey's HgxTiS2 intercalated materials , the Hg(0) atoms' 6s2 inert pairs may be forming weak coordinate covalent complexes with the sulfur 3s0 orbitals , with a slight Hg>S electron transfer . The spherical Hg 6s and S 3s orbitals have the correct shape and wave symmetry to overlap properly and form such sigma covalent bonds . These feeble Hg 6s2 >S 3s0 coordinate covalent bonds are readily broken when HgxTiS2 is heated to any significant extent .

McKelvey's studies indicated that the layered structure of the TiS2 host was maintained throughout the intercalation-deintercalation processes ; no conversion of HgxTiS2 to a thiospinel was observed in the HRTEM (High Resolution Transmisssion Electron Microscopy) and XPD (X-ray Powder Diffraction) analyses of their products . Unfortunately they didn't do any electrical conductivity/resistivity testing of their intercalated materials .

Di Salvo and co-workers (1973) prepared a series of compounds of another layered Transition metal dichalcogenide , tantalum disulfide , intercalated with various p-block elements . Mercury was readily intercalated into TaS2 from the vapor phase at 200 C , forming a stable , stoichiometric 1 : 1 adduct , Hg1.0TaS2 , in which the mercury atoms were packed into the interlayer void spaces , as in TiS2 . The intercalated products all had a hexagonal structure similar to the host TaS2 , as shown by their X-ray diffraction analyses . Again , no thiospinel structure was formed in any of the intercalated products . All of the metal–TaS2 materials were metallic solids ; several were superconductors near Absolute Zero (for Hg1.0TaS2 , Tc = 2.1 K) . Unfortunately , these researchers didn't provide any electrical conductivity/resistivity versus temperature graphs for their intercalated products , which would have been quite interesting .

Clearly , the Hg(0) in Di Salvo's Hg1.0TaS2 (and probably in McKelvey's Hg0xTiS2 as well) is electronically similar to that in elemental mercury . That is , the inert pairs are still in the 6s valence shell orbitals . However , to obtain high temperature superconductivity in zerovalent mercury compounds their inert pairs must be promoted somehow into the empty 7s orbitals and above EF , where they will be freed from the powerful electrophilic grip of the Hg2+ kernels and can become mobile and itinerant , as Cooper pairs , throughout the crystal lattice .

One way this 6s>7s promotion might be accomplished is to convert the Hg00.5MS2 intercalation compound (M = various Transition metal elements , and Sn) , with its feeble Hg0>S0 coordinate covalent bonds , into the corresponding Hg0M2S4 thiospinel , with much stronger S2>Hg2+ + 2e coordinate covalent bonds . That is , the four sulfide anions , which are powerful nucleophiles (almost as strong as cyanide anion in the Electrochemical Series) , will press down on the Hg2+ kernel , coordinating it , and “pop” the inert pairs up into the 7s orbitals .

High temperatures and pressures (HPHT) undoubtedly would be required to convert Hg00.5MS2 into Hg0M2S4 . This transformation was accomplished with CuTi2S4 , so hopefully it would also succeed with the analogous mercuryMS2 materials as well :

Hg0 (b.p. 357 C) + 2 MS2 ------ [intercalate] -----> Hg0(MS2)2 ------ [HPHT] -----> Hg0M2S4 , where M = a Transition metal (eg. Ti , Nb , Ta , Mo , W) and Sn (ie. SnS2) .

The layered , graphite-like compound MoS2 , which is a relatively inexpensive high temperature industrial lubricant (Moly) , would be particularly interesting to investigate in this regard . HgMo2S4 might prove to be an economically practical HTS material .

Tin(IV) sulfide also has a layered CdI2 structure like TiS2 , and like it has a golden yellow color and metallic appearance ; but unlike TiS2 , SnS2 is a semiconductor [ambient conductivity] . The combination of equimolar quantities of mercury and SnS2 might provide a second type of zerovalent mercury compound , HgSnS2 , with a wurtzite (or zinc blende , GIF image , 22 KB) structure :

The hexagonal wurtzite crystal structure for a binary metal sulfide . The metal cations are the smaller silvery spheres , while the sulfide anions are represented by the larger yellow spheres . This image was copied from the Wikipedia web page , Wurtzite Crystal Structure . My thanks to the author of this sketch , and Wikipedia , for implied permission to reproduce it here .

Presumably the mercury and tin atoms would be in alternating layers in the hypothetical wurtzite HgSnS2 . In the ternary wurtzite compound Cu2HgI4 , the copper(I) and mercury(II) cations are in separate layers in its room temperature form , which has a brick-red color . This ordered structure changes to a disordered one at ~ 67 C in which the Cu(I) and Hg(II) cations are randomly distributed in the cation layers . The high temperature form of Cu2HgI4 has a brown color .

The formula of copper(I) mercury(II) tetraiodide should actually be written as Cu2Hg[*]I4 , where [*] represents a tetrahedral void in the lattice . Thus , all of the components of Cu2Hg[*]I4 have tetrahedral coordinations , including the void spaces and the iodide anions . The copper cations are capable of moving through the lattice , via the voids , under an applied potential difference , making Cu2HgI4 an ionic electrical conductor .

HgSnS2 might also be prepared by the solid-state reaction of the two precursors HgS and SnS (in an HPHT apparatus to contain any mercury vapor emissions) . Mercury(II) sulfide has two crystal forms . The common one , a-HgS , occurs in Nature as the mineral cinnabar and has a chain-like structure with linear SHgS and bent HgSHg :

This image was copied from the Wikipedia web page , Mercury sulfide . Again , my thanks to the author of this sketch , and Wikipedia , for implied permission to reproduce it here .

a-HgS has an orange-red color and is (or was) often used in artists' oil paints , in particular for the vermilion pigment .

The less common form of mercury(II) sulfide , b-HgS , is black and has the cubic zinc blende crystal structure (a = 5.8517 ) . Since the mercury(II) in this form of HgS already has a tetrahedral coordination by the sulfide anions , it would be the recommended reagent for use in the preparation of HgSnS2 (and on a practical note , black HgS is considerably cheaper than red HgS , according to my Alfa-Aesar chemical catalog) .

The tin(II) sulfide precursor has a layer-like crystal structure somewhat like that of SnS2 and TiS2 :

My thanks to WebElements for the above image , which was from their web page , Tin sulphide .

In SnS both the tin and sulfur atoms have a tetrahedral sp3 hybridization , with three SnS covalent bonds and a stereochemical lone pair of electrons :

The bulky lone pairs of electrons on the tin atoms in particular extend into the interlayer spaces of the lattice , giving it a layered structure as in the Transition metal dichalcogenides .

Tin(II) sulfide is usually described as having a distorted rocksalt crystal structure (Wells) . It's interesting to compare SnS with lead(II) sulfide (the mineral galena) , which has an undistorted , genuinely cubic rocksalt structure . In the ionic picture of the chemical bonding in PbS , the lead component consists of the Pb2+ cations , surrounded by their spherical 6s2 inert pairs . The underlying Pb(IV) kernels are very powerfully oxidizing (E0red = 1.691 V) and electrophilic , and maintain a strong grip on the inert pairs , preventing them from hybridizing and forming covalent bonds . In SnS the underlying Sn(IV) kernels are weakly oxidizing (E0red = 0.151 V) , with a feeble hold on the 5s2 inert pairs . They can hybridize to form covalent bonds , clearly present in the tin(II) sulfide lattice , which is only superficially similar to the rocksalt structure in the alternating arrangement of its tin and sulfur atoms .

In the solid state reaction of b-HgS and SnS the Sn(II) will be oxidized by the Hg(II) :

Sn(II) – 2e -------> Sn(IV) ; E0ox = – 0.151 V ;

Hg(II) + 2e ----> Hg(0) ; E0red = 0.851 V ;

Net reaction : Hg(II) + Sn(II) -------> Hg(0) + Sn(IV) ; E0T = 0.7 V .

The Hg(II) will in effect seize the tins' 5s2 lone pairs ; but will those electrons become mercury's 6s or 7s inert pairs ? Since the mercury atoms in the b-HgS precursor reagent are already tetrahedrally surrounded by four sulfide ligands , hopefully the transferred electrons will be sterically blocked by them and will occupy the empty 7s frontier orbitals above the Hg2+ kernels :

b-HgS (m.p. 850 C*) + SnS (m.p. 882 C) -------- [HPHT] -------> HgSnS2 (wurtzite) .

* some references give m.p. 583.5 C (sublimes) ; maybe this is for a-HgS , the red cinnabar form ?

Perhaps the simplest synthesis of HgSnS2 might be the direct combination of its three component elements in a high pressure reactor (to contain the mercury vapor) :

Hg0 (b.p. 357 C) + Sn0 (m.p. 232 C) + 2 S0 (m.p. 115 C , b.p. 445 C) -------- [HPHT] -------> HgSnS2 .

In a variation of this method , the mercury and tin could first be combined to form the amalgam HgSn (an alloy , or maybe an intermetallic compound) . HgSn is a known material ; it's a superconductor near Absolute Zero (Tc = 4.20 K) . One equivalent of HgSn would be finely granulated in an agate mortar and triurated again with two equivalents of sulfur powder . The HgSn + 2 S0 mixture would then be cooked to HgSnS2 in a HPHT apparatus .

Remarkably , this very reaction was first investigated by alchemists back in the Middle Ages . They were as usual trying to convert tin or mercury into gold , but instead obtained mosaic gold , the artist's term for tin(IV) sulfide . As mentioned above , SnS2 has a golden-yellow color and metallic luster , and is sometimes used by artists as a gold flake” pigment . Quoting from the Wikipedia article about mosaic gold :

“Alchemists prepared this [SnS2] by combining mercury , tin , sal ammoniac [ammonium chloride] , and flowers of sulfur , grinding , mixing , then setting them for three hours in a sand heat . The dirty sublimate being taken off , aurum mosaicum was found at the bottom of the matrass”.

Under such conditions the volatile mercury undoubtedly distilled out of the mixture during the heating stage (in the “dirty sublimate”) and was lost in the air . Nevertheless , it's somewhat amusing (and ironic) to speculate that the alchemists may have produced in the “aurum mosaicum” phase small amounts of HgSnS2 which may be a high temperature superconductor many centuries ago !

A Valence Bond electron distribution in HgSnS2 is sketched as follows :

The sketch illustrates the formation of the SnS covalent , S2>Hg2+ coordinate covalent , and Hg 7s XO metallic bonds in the hypothetical wurtzite compound . The mercury 6s2 inert pairs have (optimistically) been promoted to the empty 7s frontier orbitals , under pressure from the nucleophilic sulfide anions tetrahedrally coordinating the electrophilic Hg2+ kernels .

A similar sort of “element cooking” reaction could be tried for the various other Transition metals that are known to form layered dichalcogenides , for example , Ti , Nb , Ta , and Mo . However , these latter elements may not readily (if at all) form wurtzite or zinc blende compounds , as they usually prefer at least octahedral coordinations by anions and other ligands . Titanium does have a tetrahedral coordination in the familiar molecule TiCl4 , but I don't know of any titanium-based wurtzite or zinc blende compounds , which are generally associated with p-block elements .

Mercury(0) might also be inserted into the lattice of silicon disulfide , SiS2 , first forming an intercalated compound like those of the Hg0(MS2)x series (M = Ti , Nb , Ta , Mo , W and Sn) discussed above . Unlike the layered MS2 host lattices with their graphite-like structures , remarkably SiS2 is an inorganic polymer with a fibrous morphology and is composed of long SiS2 chains with substantial interchain void spaces :

This image was copied from the Wikipedia web page , Silicon sulfide . Again , my thanks to the author of this sketch , and Wikipedia , for implied permission to reproduce it here .

Using equimolar proportions of Hg(0) and SiS2 , the intercalated product Hg0(SiS2)1.0 could first be formed . It would then recrystallize in the more compact wurtzite (or zinc blende) Hg0SiS2 structure when heated and compressed under HPHT conditions :

Hg0 (b.p. 357 C) + SiS2 (m.p. 1090 C , subl.) ------ [intercalate] -----> Hg0(SiS2)1.0 -------

------ [HPHT] -----> Hg0SiS2 .

Silicon disulfide is commercially available (eg. Alfa-Aesar) at a moderate cost . It's a thermally stable white solid which is very moisture-sensitive , hydrolysing rapidly to silica when exposed to atmospheric humidity (SiS2 has been described as having a “rotten egg odour” from H2S evolved from its surface) . It should be handled in a glove box in bone-dry air .

Since SiS2 can be synthesized by heating together silicon and sulfur powders (see this U.S. patent) , it might be possible to prepare Hg0SiS2 directly from its three elements . Mercury metal , very fine silicon powder , and flowers of sulfur would be ground together in a 1 : 1 : 2 molar ratio in a mortar to make a black , solid “premix . This mixture would then be “cooked” under HPHT conditions in an anvil press to form Hg0SiS2 as the wurtzite or zinc blende compound :

Hg0 (b.p. 357 C) + Si0 (m.p. 1414 C) + 2 S0 (m.p. 115 C , b.p. 445 C) --------

[grind together in mortar] -------> premix -------- [anvil press , HPHT] -------> Hg0SiS2 .

Unlike the Transition metal elements (and tin) in the Hg0M2S4 series of normal spinels , silicon can't be octahedrally coordinated with sulfide under these conditions , so the intercalated compound Hg0(SiS2)2.0 couldn't be converted into the hypothetical normal spinel Hg0Si2S4 , even in an anvil press under HPHT conditions . Silicon is octahedrally coordinated by fluorine atoms in the discrete hexafluorosilicate anion , SiF62- (as in the salt MgSiF6) , but I doubt that silicon will octahedrally coordinate with sulfur at all .

Hg0SiS2 should resemble the compound Hg0SnS2 discussed above , with all the atoms tetrahedrally coordinated to each other and the Hg(0) 6s2 inert pairs promoted up into the 7s frontier orbitals . The sulfides should transfer some of their charge to the Hg2+ kernels , thus effectively cancelling their high reduction potential . The promoted inert pairs should then be stabilized , hopefully as Cooper pairs in the 7s sigma XO metallic bond , thereby making Hg0SiS2 an excellent HTS material .


TinSulfur HTS Candidate Compounds


The inert pair effect occurs in other heavy p-block elements , as shown in the tabulation near the top of this web page . The strength to which the ns2 inert pairs are bonded to their respective ns0 kernels is roughly measured by the standard oxidation potential , E0ox , of the ns2 species ; or conversely , by the standard reduction potential , E0red , of the ns0 species . The E0red of Hg2+ at + 0.851 V is a “medium range” value ; several other inert pair elements have considerably higher reduction potentials , such as Tl(III) , Pb(IV) , and Bi(V) , as shown in the inert pairs table above .

But several elements have lower redox potentials than Hg , and that makes them very interesting in this present context of converting the inert pairs into Cooper pairs in HTS materials . Tin is particularly attractive in this respect . Tin(II) is readily oxidized (E0ox = – 0.151 V) to tin(IV) , which correspondingly is a mild oxidizer (E0red = 0.151 V) . It should therefore be considerably easier to “pop” the tin 5s2 inert pair up into the next higher energy level , 6s , than promoting the mercury(0) 6s2 inert pair to 7s . How might this be accomplished chemically ?

The synthesis of the metallic solid tin(III) phosphide , SnP , by Donohue in 1970 demonstrates how this could be done , and rather simply , too (although the researcher must have the suitable apparatus to carry out high pressurehigh temperature reactions , that is , in an “anvil” type of press) . SnP was prepared by the combination of equimolar quantities of tin metal powder and red phosphorus in a tetrahedral anvil press under an applied pressure of 65 kbars (~ 65,000 atm) and at 800 C [also at 65 kbars and 1200 C , and at 15 kbars and 900 C] . Two forms of SnP , both having high electrical conductivities , were produced in Donohue's experiments . The tetragonal form displayed the crystallographic bulging” suggestive of the voluminous , stereochemical tin(II) 5s2 inert pairs . That is , this material contained partially disproportionated tin(II) and (IV) . The cubic rocksalt form of SnP apparently had fully reproportionated Sn(III) , i.e. Sn(IV) + e . Unlike the tetragonal form , it was able to become superconducting somewhere in the range of 2.8 and 4.0 K .

A Valence Bond analysis of the electron distribution in cubic SnP is sketched as follows :

The tin and phosphorus atoms are surrounded by an energetically stable twelve-set” of valence shell electrons in an octahedral configuration , thus resulting in the cubic rocksalt crystal structure of the material . Such twelve-sets require the use of hypervalent [outside of the normal valence shells] orbitals and electrons . In order to obtain enough electrons for its twelve-set , tin mobilizes two of its 4d10 orbitals (probably the higher energy x2–y2 and z2 ones) , with their contained four electrons , combining them with its four normal valence electrons in the d2sp3 octahedral hybrid orbital .

The 4d orbitals in tin are at about the same energy level as the 5s and 5p valence shells (GIF image , 16 KB) . The 5s + 4d + 5p orbitals and their electrons can be hybridized fairly easily , with relatively small promotion energies involved . In fact , white tin metal may be using its 4d orbitals and electrons in d2 + p2 + spz distorted octahedral orbitals (GIF image , 31 KB) , which form the zigzagging Sn–Sn covalent bonds (GIF image , 35 KB) in the bulk metal .

Meanwhile , phosphorus , as a p-block element , will combine two of its outer , empty 4p hypervalent orbitals with its normal 3s-p orbitals to form the sp5 octahedral hybrid orbital , in which its five valence electrons are located . The tin d2sp3 orbitals with seven electrons per Sn atom are combined with an equal number of phosphorus sp5 orbitals with five electrons per P atom to form the three-dimensional rocksalt lattice , with six Sn–P covalent bonds per atom .

The formation of hypervalent sp5 hybrid orbitals apparently requires a very large promotion energy that can be provided , for example , by exoergic fluorination reactions . Such seems to be the case with the chemically unreactive gas sulfur hexafluoride , SF6 , which was discussed in the Valence Bond web page . A similar situation probably occurs with the unreactive gas phosphorus pentafluoride , and in the hexafluorophosphate anion , (PF6). Clearly , sufficient energy required by the phosphorus atoms to create hypervalent sp5 hybrid orbitals and use them in the rocksalt SnP lattice is being provided by the high temperatures of the Sn + P reaction in the anvil press .

Recall that tin had mobilized eight electrons (4d4 + 5s2 + 5p2) . What happened to the unused eighth electron ? There's no room in the normal valence shells for it , as they have been “taken” by the Sn–P covalent bonds . The “extra” eighth electron has been “popped” up to the empty 6s frontier orbital . These voluminous , now half-filled orbitals can overlap continuously throughout the SnP lattice to form a sigma XO metallic bond . Indeed , SnP has an excellent electrical conductivity over a wide range of temperatures (50,000 ohm-1-cm-1 at 25 C , and 333,000 ohm-1-cm-1 at 4.2 K) . And as mentioned , the cubic form becomes superconducting at ~ 4 K .

We now see how it might be possible to chemically convert tins' 5s2 inert pairs into Cooper pairs in a practical solid state system . Donohue's SnP preparation could be repeated , but this time substituting sulfur for the phosphorus to obtain SnS instead . And , under HPHT conditions in an anvil press , the cubic form of SnS should form , rather than the ordinary” layered form of tin(II) sulfide that is presently known (sketch above) . Since sulfur has six valence electrons – one more than phosphorus – two of tin's electrons (its 5s2 inert pair) would be “popped” into its 6s frontier orbitals .

The Valence Bond sketch of this hypothetical metallic , cubic form of SnS would resemble that of SnP above , now with two tin valence electrons in the 6s orbital . However , VB warns us that an alternate non-metallic , insulating cubic form of tin(II) sulfide could also theoretically form :

In this latter case , only one of tin's hypervalent 4d orbitals , with two electrons , needs to be drafted for the required twelve-set . The dz2pz + spx + spy composite hybrid orbital , while theoretically possible , would be most unusual . This is the first time I have encountered such a hybrid orbital in my extensive experience using the Valence Bond theory and sketches to describe covalent bonds in inorganic solids . Theoretical considerations aside , the simple fact is that a system with an even number of valence electrons tends to use them in pairs for the formation of the strong , low energy covalent bonds in that system . That is , the atoms in it will be arranged so that all of the pairs are used in the covalent – not metallic – bonds . The reader has undoubtedly heard the expression , “Nature abhors a vacuum” ; it can equally be said that Nature abhors a metallic bond , and tries every combination of atomic coordinations in a solid state structure to get rid of such a high energy aberration !

The d2sp3 octahedral hybrid orbital would be energetically more favorable than the alternate dz2pz + spx + spy composite hybrid orbital , since it uses the readily accessible 4d4 + 5s2 + 5p2 orbitals and electrons . The dz2pz + spx + spy hybrid orbital would be obliged to use the higher energy 6s native orbital in its construction , which I'm not sure is actually possible . So a metallic form of cubic SnS would be favored by VB over its rival nonmetallic form , but I suspect Nature would still somehow manage to avoid metallic SnS anyway it could !

Two alternate forms of high pressure SnS could conceivably be the cubic rocksalt ionic Sn2+S2, the tin analogue to galena (this would be nonmetallic or semiconducting , like PbS) ; and cubic zinc blende/sphalerite , or hexagonal wurtzite SnS , with tetrahedrally coordinated Sn and S atoms . These latter forms of SnS would be metallic ,with “popped” 5s2 inert pairs , and possibly HTS materials .

That said , the HPHT combination of equimolar quantities of tin metal powder and flowers of sulfur in an anvil press to synthesize cubic SnS would be a very worthwhile experiment , leading as it might to the production of a new metallic solid and maybe a high temperature superconductor . The researcher should also be aware , though , that the resulting cubic SnS could be a nonmetallic , insulating compound .

The above considerations can be generalized to design analogue binary compounds , also possibly metallic and maybe HTS materials , based on the promotion of ns2 inert pairs in p-block heavy metal elements to (n+1)s0 frontier orbitals . The idea comes from semiconductor chemistry , which involves binary MX compounds having a zinc blende (sphalerite) / wurtzite crystal structure , with 4 : 4 tetrahedral coordinations of the constituent M and X atoms . They are selected so that their combined valence electrons add up to 8 , the electronically stable octet ; that is , four pairs of valence electrons in the MX covalent (or coordinate covalent , or ionic) bonds in the compounds . Three combinations are numerically possible : 1 + 7 = 8 ; for example , Cu(1) + Cl(7) = CuCl(8) ; 2 + 6 = 8 ; for example , Zn(2) + S(6) = ZnS(8) ; and 3 + 5 = 8 ; for example , Ga(3) + As(5) = GaAs (8) . The 4 + 4 = 8 combination is conceivable , but I can't think of an example of it here .

Binary MX compounds having promoted ns2 inert pairs could be similarly designed , but setting aside the ns2 valence electrons in reserve , so they can be “popped” in the subsequent MX HPHT synthesis . For example , tin has 5s2 5p2 valence electrons . Its 5s2 inert pair is set aside , leaving the 5p2 valence electrons for the octet (zinc blende/wurtzite) or twelve-set (cubic rocksalt) . That is , tin will have a “combining power” of 2 in the MX syntheses . Germanium (4s2 4p2) would be similar . Meanwhile , sulfur has six usable valence electrons (3s2 3p4) . Thus , we have tin(2) + sulfur(6) = SnS(8) [zinc blende/wurtzite] + the “popped” inert pair in the 6s frontier orbital . Selenium (4s2 4p4) and tellurium (5s2 5p4) would be analogous to sulfur . The various MX combinations are tabulated in the following sketch :

If a cubic rocksalt MX compound is desired , the M atom (with the inert pair) must be able to use two of its nd10 orbitals (with their four electrons) in the formation of the hypervalent twelve-set of electron pairs around the octahedrally-coordinated M and X atoms . Note that the three heaviest of the p-block element cations with inert pairs , Tl1+, Pb2+, and Bi3+, have been omitted from the above tabulation . Their very high oxidation potentials (– 1.252 V , – 1.691 V , and – 1.759 V , respectively) suggest that their 6s2 inert pairs are too strongly bound to their respective kernels , and therefore probably couldn't be “popped” into their 7s frontier orbitals , even under very forcing , energetic HPHT conditions . See , however , the section on the HP–HT cubic perovskite LaBiS3 , below . Maybe .....

Another way to “pop” the inert pairs in M might be to blend the M atoms into an existing host structure AX having tetrahedrally or octahedrally A and X coordinated atoms . MX will then hopefully adopt the same structure as the host AX lattice .

The hypothetical compound SnZnS2 illustrates this approach . ZnS can have either the zinc blende / sphalerite [the mineral form of ZnS] or wurtzite crystal structure ; in any case , the zinc and sulfur atoms in it always have a 4 : 4 tetrahedral coordination . If SnS is created within ZnS , its Sn and S atoms should mimic the host coordinations as well , so as to smoothly mix into the structure . In “ordinary” SnS the tin and sulfur atoms both have tetrahedral coordinations , although with a ligancy of three , plus their respective inert pairs . In SnZnS2 their inert pairs would disappear , and they would have an increased ligancy of four Sn–S covalent bonds . Tin's 5s2 inert pairs should be promoted up into its 6s frontier orbitals in the new structure :

SnZnS2 might be prepared by a direct combination of the three elements concerned :

Sn0 (m.p. 232 C) + Zn0 (m.p. 420 C) + 2 S0 (m.p. 115 C , b.p. 445 C) -------- [HPHT] -------> SnZnS2 .

Note that the reaction of zinc dust and flowers of sulfur is highly exothermic , and the mixture can detonate like gunpowder (the tin might moderate it somewhat , though) . Zinc could be used to reduce the tin(IV) in the layered compound SnS2 (mentioned above in connection with the Hg0M2S4 spinels) in a more tranquil reaction :

SnS2 (m.p. 600 C , dec.) + Zn0 (m.p. 420 C) -------- [HPHT] -------> SnZnS2 .

The combination of tin powder , flowers of sulfur , and zinc sulfide might be an efficient , economical , and safe method :

Sn0 (m.p. 232 C) + S0 (m.p. 115 C , b.p. 445 C) + ZnS (m.p. 1700 C) -------- [HPHT] -------> SnZnS2 .

In the HPHT conditions the SnS (m.p. 882 C) would form as an intermediate phase ; it would act as a high temperature inorganic solvent (flux) , dissolving the the refractory ZnS (m.p. 1700 C) . As the reaction mixture is cooled to STP , the SnS would co-crystallize with the ZnS in alternate layers , adopting its wurtzite (or zinc blende) structure . The Sn(II) 5s2 inert pairs would be popped off the Sn(IV) kernels and promoted up into the 6s frontier orbitals . Hopefully the SnZnS2 composite material would be a double wurtzite, like Cu2HgI4 (changes to a disordered) , with alternating layers of tetrahedral Sn(IV)S and ZnS . And , of course , with the Cooper pairs in the metallic Sn–S layers .

The zinc chalcogenides (S , Se , and Te , and ZnO) all have a wurtzite (or zinc blende) crystal structure . The theoretical problem of nonmetallic , insulating cubic rocksalts and the M element's hypervalent nd10 orbitals and electrons wouldn't arise in these zinc blende / wurtzite lattices , which utilize only sp3 tetrahedral orbitals in their construction .

Tin(II) might have a tetrahedral coordination by sulfide anions in a thiospinel , Atet(Moct)2S4 , as mercury(0) should in the Hg0M2S4 compounds discussed above . A very small trivalent cation that strongly favors an octahedral coordination would be selected as the Moct component . Al3+ should be suitable , so the target spinel would then be SnAl2S4 [SnIVtet(2e)(Al3+oct)2S4 ] :

Sn0 (m.p. 232 C) + 2 Al0 (m.p. 660 C) + 4 S0 (m.p. 115 C , b.p. 445 C) -------> SnAl2S4 .

Since the reaction of aluminum and sulfur is very exothermic , a milder variation of it might employ the commercially available reagent Al2S3 :

Sn0 (m.p. 232 C) + S0 (m.p. 115 C , b.p. 445 C) + Al2S3 (m.p. 1100 C) -------> SnAl2S4 .

As in the proposed synthesis of SnZnS2 , the SnS formed in situ could act as a flux to dissolve the more refractory Al2S3 . The normal thiospinel SnAl2S4 would hopefully crystallize from the melt as it is slowly cooled to room temperature .

A third possible synthesis method for SnAl2S4 would involve the solid state metathesis reaction of the tin(II) and Al(III) halides with a suitable metal sulfide . All three reagents must be anhydrous , as the sulfides are sensitive to hydrolysis to their respective oxides (plus the highly toxic H2S) :

SnCl2 (m.p. 247 C , b.p. 623 C) + 2 AlCl3 (m.p. 193 C , 180 C subl.) -------> SnAl2Cl8 [uncharacterized intermediate] ------ [4 Li2S (m.p. 1372 C)] -------> SnAl2S4 + 8 LiCl (m.p. 610 C , could act as a flux for the reaction) .

The reaction mixture of SnAl2S4 + 8 LiCl would be thoroughly extracted with anhydrous methanol (b.p. 65 C) in a Soxhlet apparatus (GIF image , 32 KB) to remove the LiCl (solubility 423.6 g/L at 25 C in MeOH) . Some solid state reactions are quite exothermic , and the by-product LiCl might melt and act as a flux in the reaction mixture . However , the sulfide product might still need some HPHT annealing to ensure its complete conversion into the normal thiospinel with tins' 5s2 inert pairs “popped” into the 6s orbitals .

Germanium might form analogous compounds , GeZnX2 (X = S , Se , Te) . Germanium has a slightly lower intervalence redox potential [Ge(II) – 2e ----> Ge(IV) ; E0ox = 0 V] than Sn(II) , so its 4s2 inert pairs should be very easy to promote to its 5s0 frontier orbitals . Also , germanium shows a strong preference for a tetrahedral coordination in its compounds . That coordination should be evident in these proposed wurtzite compounds , which might be synthesized by element cooking in an HPHT apparatus :

eg. Ge0 (m.p. 938 C) + Zn0 (m.p. 420 C) + 2 Se0 (m.p. 221 C , b.p. 685 C) ---- [HPHT] ----> GeZnSe2 .

Such germanium compounds would really be of academic interest only , as germanium is a rare , costly element , and its compounds are correspondingly very expensive .

Germanium's lighter analogue in the carbon (IVB/14) family , silicon , is one of the commonest elements in the Earth's crust , and is a very cheap and abundant industrial commodity :

A small chunk of industrial grade silicon in the author's collection . Germanium (pictures) has a similar shiny , metallic appearance , but lacks the bluish color of silicon .

The Si(II) to Si(IV) redox intervalence transition is probably comparable to that for germanium and tin , but the Si(II) 3s2 inert pairs would have to be promoted up into the 4s frontier orbitals . There is a much larger energy gap between the 3s-p to 4s-p levels than with the narrower 4s-p to 5s-p and 5s-p to 6s-p levels (GIF image , 16 KB) . Very likely this forbidding energy gap would prevent the promotion of silicon's 3s2 inert pairs to the 4s-p level , thus ruling out any metallic silicon-based compounds .

The attempted reproportionation of aluminum(0) metal and aluminum(III) oxide was unsuccessful in that the desired lower-valent aluminum(I) oxide couldn't be isolated at room temperature . There is some evidence it was briefly stable at very high temperatures , but promptly disproportionated to the starting materials as the reaction mixture was cooled down . The aluminum 3s2 inert pairs in the ephemeral Al2O could not be promoted up into the 4s frontier orbitals ; the 3s-p to 4s-p energy gap was simply too wide for them to cross . That made the Al2O thermodynamically unstable and susceptible to disproportionation .

From the above considerations we see that two general constraints would be imposed on the viability of any proposed new metallic compounds based on promoted inert pairs :

* The inert pairs must come from the 4s , 5s , and 6s valence shells (degenerate atoms) . The 2s inert pairs are absolutely ruled out from consideration , and 3s inert pairs probably couldn't be promoted either ; and second ,

* The redox intervalence transition between the nsx and nsx2 species must not be too high (on the oxidizing side) , but preferably should be as low as possible . Mercury's redox potential of 0.851 V is probably the upper limit of what might be feasible in actual laboratory practice . Those higher potentials for Tl(III) , Pb(IV) , and Bi(V) suggest that the 6s2 inert pairs in Tl1+ , Pb2+ , and Bi3+ are too strongly bound to their respective highly electrophilic kernels , and cannot be promoted up into their 7s frontier orbitals , even though the 6s-p to 7s-p energy gap is relatively narrow .

In the search for suitable p-block elements with inert pairs a compromise must therefore be sought between a fairly high energy level inert pair , combined with a low redox intervalence transition between its nsx and nsx2 species . An acceptable compromise is provided by the element tin , whose 5s2 inert pairs should be easy to promote to the 6s-p level , and which has a small intervalence transition redox potential (0.151 V) between its Sn(II) and Sn(IV) species . Additional encouraging benefits provided by tin include a very low toxicity of tin metal and inorganic tin compounds , and a relatively modest cost for tin and its compounds , which are readily available as industrial and fine chemical commodities .


The Cubic Perovskite LaBiS3 : A Possible HTS Material


The BiS2 family of layered superconductors has been extensively explored in the last few years . They are somewhat similar in crystal structure to the LaOFeAs ferropnictides , with alternating covalentmetallic BiS and ionic MxXy layers (M = an Alkaline Earth or Rare Earth cation , and X = an oxide or fluoride anion) . The structure of SrFBiS2 is typical of that of the BiS2 compounds :

My thanks to the author and/or copyright holder of the original image , which I have annotated slightly .

The bismuth atoms in this material are clearly in the trivalent state , and their stereochemical 6s2 inert pairs (not shown in the sketch) are located in the interlayer spaces . SrFBiS2 is a semiconductor with a direct electrical conductivitytemperature relationship (H. Lei and co-workers , 2012) . Other members of the BiS2 family have been found to be True Metals (with an inverse electrical conductivitytemperature relationship , ie. metallic) and superconducting , but only very near to Absolute Zero .

Suppose a third formula sulfur atom was added to the BiS2 , to obtain BiS3 , with an octahedral coordination of the sulfurs . Could an ABiS3 perovskite be synthesized in which the bismuth 6s2 inert pairs would be strongly squeezed by the sulfur atoms and popped” into the vacant 7s frontier orbitals ? The only practical A trivalent cation suitable for such a perovskite would be La3+, in the compound LaBiS3 . If this material had a cubic perovskite crystal structure , it could conceivably be metallic and , with its 6s2 inert pairs now relocated above EF in the 7s orbitals , it could also be a high temperature superconductor . If its Cooper pairs were as strongly associated as the original inert pairs were in the Bi(III) precursor , LaBiS3 might even be an ambient superconductor ! The following sketch illustrates a suggested valence electron distribution in LaBiS3 :

Contributions to the BiS twelve-set are : 3 electrons from La0 , which becomes La3+ ; the 3 6p3 electrons from Bi(III) [the 6s2 inert pair has been popped” up to 7s] ; and 2 electrons from each of the three formula sulfur atoms , or 6 total . That is , La(3) + Bi(3) + S3 (6) = 12 . The sp5 octahedral hybrid orbital (spz + pxy2 + pxy2) , typical of the p-block elements , holds the twelve-set . In a very energetic environment , the sp5 hybrid orbital will “take” the 6s orbital for its own use and displace the inert pair into the empty 7s frontier orbital . Such energetic conditions will have to be supplied by an HP–HT synthesis of LaBiS3 in an anvil press :

La2S3 (m.p. 2110 C) + Bi2S3 (m.p. 850 C) ---- [HPHT] ----> LaBiS3 .

The lower-melting Bi2S3 might act as a flux to dissolve the more refractory La2S3 . The cubic perovskite LaBiS3 would hopefully crystallize as the reaction mix is cooled from its peak temperature of ~ 900 C (say) to room temperature , with the high pressure maintained until ambient temperature is reached .

Alternately , the three parent elements might be combined directly in the proper proportions :

La0 (m.p. 920 C) + Bi0 (m.p. 271 C) + 3 S0 (m.p. 115 C , b.p. 445 C) ---- [HPHT] ----> LaBiS3 .

Using the following values for the crystal ionic radii (per Shannon and Prewitt) of the LaBiS3 components , the tolerance factors for LaBiIIIS3 and LaBiV(e22)S3 can be calculated : La3+ , 12-coordinate , r = 1.36 ; Bi(III) , 6-coordinate , r = 1.03 ; Bi(V) , 6-coordinate , r = 0.76 ; and sulfide anion , 2-coordinate (linear) , r = 1.59 (estimated from 6-coordinate , r = 1.84 , by taking a ratio with octahedral and linear oxides) . For LaBiIIIS3 , t = 0.71 , which is well below the limit for perovskites ; LaBiIIIS3 might have the ilmenite crystal structure . For LaBiV(e22)S3 , t = 0.89 , which is within the range of tolerance values for cubic symmetry perovskites . This analysis suggests that if the bismuth inert pair can be promoted up into the 7s orbitals , the sulfur atom ligands can coordinate the bismuths as Bi(V) , resulting in LaBiS3 having the stable cubic perovskite structure . However , very forcing HPHT conditions must be used to obtain LaBiV(e22)S3 . An ambient pressure ceramic (shake-'n-bake) synthesis would undoubtedly result in the production only of the electronically-inert LaBiIIIS3 .

As discussed above , the very electrophilic Bi(V) kernel has a powerful grip on the inert pair , as implied by its high reduction potential to Bi(III) , E0red = 1.759 V . A combination of the “chemical pressure” of the six coordinating nucleophilic , charge-donating sulfur atoms forcing down on each bismuth atom , and of the mechanical pressure applied by the anvil press squeezing the atoms together , might be enough to promote the inert pairs into the 7s orbitals . If that can be accomplished , LaBiS3 should be metallic and probably superconducting , maybe at a remarkably high temperature .

It would be very interesting in this context to compare the oxide analogue , LaBiO3 , with LaBiS3 . Using the known oxide crystal ionic radius of r = 1.21 (per Shannon and Prewitt , for a linear , 2-fold coordination) , the tolerance factor for LaBi(III)O3 is calculated as t = 0.81 , while that of the metallic LaBi(V)(e22)O3 is t = 0.92 . Again , the latter more compact cubic perovskite form would be energetically favoured under the very high pressure conditions experienced in an anvil press synthesis procedure .

However , this time LaBi(V)(e22)O3 might be metastable only at high pressures , and transform to the electronically insulating LaBi(III)O3 when the pressure is released . Unlike the chemically reducing sulfides in LaBiS3 , the oxides in LaBiO3 are redox inert , neither oxidizing nor reducing in nature . Therefore in LaBiO3 the underlying Bi(V) kernels remain very powerfully oxidizing . The electron gas (the promoted 6s2 inert pairs) could infiltrate back through the lattice to their parent Bi(V) kernels and bond to them again as stereochemically active inert pairs in LaBi(III)O3 under ambient conditions . The prediction , then , is that LaBi(V)(e22)O3 would probably be unstable and transform to LaBi(III)O3 when cooled to room temperature , while LaBi(V)(e22)S3 should be quite stable (and metallic , and possibly superconducting) at STP . The synthesis and study of LaBiS3 and LaBiO3 would be a highly worthwhile project !


Concluding Remarks


Various superconductors have in the past incorporated heavy p-block metal elements in their composition , as the electronically-active components of those materials . Such superconductors include , for example , Ba0.6K0.4BiO3 (Tc = 29.8 K) , BSCCO-2212 (Bi2Sr2Ca2Cu3O10+x , Tc = 110 K) , Tl2Ba2Ca2Cu3O10 (Tc = 122 K) , and Hg-1223 (HgBa2Ca2Cu3O7-8 , Tc = 133 K) . In a sense , these earlier superconductors might have benefitted from the inert pair enabling mechanism for HTS . It should be carefully noted that in the above formulas the heavy metal has a NIOS (non-integral oxidation state) valence , and that the four materials cited are all Robin-Day Class II mixed-valent compounds . They were all prepared under very strongly oxidizing conditions .

The resonating singlet electrons in their metallic bonds may have been predisposed to reform their respective inert pairs , which as we have seen above is a powerful thermodynamic driving force in the p-block heavy metal elements . However , when the inert pairs recombine , they do so above EF , thus forming the Cooper pairs required for superconductivity . This may at least partially explain the unusually high transition temperature (~ 30 K) for the mixed-valent bismuthate , Ba0.6K0.4BiO3 . A similar situation occurs in the other three HTS compounds . However , their resonating singlet electrons might have received additional assistance from the AFM (antiferromagnetic) induction enabling mechanism (GIF image , 77 KB) . Again , it should be carefully noted that the copper atoms in the three compounds are homovalent Cu(II) : not mixed-valent (as some people think) . The CuO layers in the materials are strongly AFM , and induce an antiparallel ordering in the metallic bond free electrons above EF , which are otherwise Pauli paramagnetic and have random spin orientations . The AFM-induced antiparallel ordering assists them in condensing into Cooper pairs at temperatures considerably higher than without the induction (as with Ba0.6K0.4BiO3 : no CuO) . The inert pair enabling mechanism further assists their free singlet electrons in recombining into the energetically stable inert pairs , now as Cooper pairs .

In contrast , the compounds proposed in this report are all homovalent , with IOS (integral oxidation state) valences (the mercury ones are zerovalent) , and don't rely on any sort of mixed-valent resonance for the functioning of their metallic bonds . They could be considered as reducing systems , given the chemically reducing nature of their chalcogen (eg. sulfide) anions . In fact , they should be carefully protected from air oxidation during their syntheses . The critical objective is to maintain the inert pairs intact , to as high a temperature as possible , and not to smear them out in a mixed-valent resonance . All these considerations result in a completely different approach to the preparation these new types of inert pair-containing metallic solids .

If it could be experimentally confirmed , the inert pair promotion effect would be another valuable enabling mechanism for high temperature superconductivity . As such , it deserves a careful investigation and verification by solid state chemists so it may take its place in the repertoire of superconductor design and synthesis .


References and Notes


inert pair effect : A.R. West , Basic Solid State Chemistry , John Wiley , New York (1988) ; pp. 106-107 ; idem. , Solid State Chemistry and Its Applications , John Wiley , Chichester (UK) , 1984 ; pp. 314-315 . Inert pairs of electrons in inorganic compounds are similar to the lone pairs of electrons in both inorganic and organic compounds (for example , the two lone pairs on the oxygen atom of the water molecule) . The presence of inert pairs in a crystal structure is a reliable diagnostic of covalent bonding in it . Sidgwick discusses in some detail the anomalous properties of elemental mercury and the inert pair effect : N.V. Sidgwick , The Chemical Elements and Their Compounds , Vol. 1 , Clarendon (Oxford University) Press , London (UK) , 1951 ; pp. 285-288 . This comprehensive inorganic chemistry textbook can be downloaded for free from the library resources web page [PDF , 11,921 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 .

which are extremely inert : E.K.S. Liu and R.J. Lagow , “The Preservation of Metal–Carbon Bonds and Metalloid–Carbon Bonds During Direct Fluorination : A Surprise Even to Fluorine Chemists”, J. Amer. Chem. Soc. 98 (25) , pp. 8270-8271 (1976) . Dimethylmercury , H3C–Hg–CH3 , was perfluorinated at – 78 to – 90 C to bis(trifluoromethyl)mercury in a 6.5% yield . F3C–Hg–CF3 can be further fluorinated at 0 C and at room temperature to CF4 and HgF2 . The authors interpret these findings as demonstrating the surprising strength of the Hg–C bonds . However , it should be noted that the fluorine attacked the carbon and hydrogen atoms in dimethylmercury , and not the mercury atom . That is , the fluorine failed to oxidize the mercury 5d10 electrons , which seem to be deeply embedded in the mercury kernel and inaccessible even to fluorine , the most powerful oxidizer known in chemistry .

In contrast , the Pt(II) in K2Pt(CN)4 (5dz2 electrons) can be easily oxidized by aqueous bromine at ambient temperature to produce the partially-oxidized salt K2Pt(CN)4Br0.3 . 3H2O , which crystallizes in the form of long , slender , needles with a coppery color , metallic luster , and having a modest electrical conductivity of ~ 830 ohm-1cm-1 at room temperature . This is the synthetic metal “KCP”, which I've written about in several Chemexplore web pages (eg. the Solids web page) . Liu and Lagow's research cited above implies that an analogous reaction could not be carried out with dimethylmercury to convert it into a synthetic metal like KCP . Mercury's 5d10 electrons are simply too inert for any sort of oxidizer , which instead will attack the carbon and hydrogen atoms in the organomercury compound .

The 3d10 electrons in zinc and the 4d10 electrons in cadmium are similarly inert and cannot be used to form hypervalent twelve-sets . Thus , Zn , Cd , and Hg never have octahedral covalent M–X bonds (CdO has the cubic rocksalt crystal structure , but its bonds are ionic , not covalent) . Their covalent bonding is restricted to the s-p orbitals : sp linear and sp3 tetrahedral .

s-p conduction band : A.B. Ellis et al. , Teaching General Chemistry , A Materials Science Companion , American Chemical Society , Washington , D.C. , 1993 ; see Figure 7.2 , p. 189 , showing simple band structures for sodium and magnesium metals .

French researchers : 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) .

McKelvey and co-workers : E.W. Ong , M.J. McKelvey , G. Ouvrard , and W.S. Glaunsinger , “Mercury Intercalates of Titanium Disulfide : Novel Intercalation Compounds”, Chem. Mater. 4 (1) , pp. 14-17 (1992) ; M. McKelvey , R. Sharma , E. Ong , G. Burr , and W. Glaunsinger , “Dynamic Atomic-Level Observation of Staging Phenomena in Silver and Mercury Intercalates of Titanium Disulfide”, Chem. Mater. 3 (5) , pp. 783-786 (1991) ; M. McKelvy , M. Sidorov , A. Marie , R. Sharma , and W. Glaunsinger , “Dynamic Atomic-Level Investigation of Deintercalation Processes of Mercury Titanium Disulfide Intercalates”, Chem. Mater. 6 (12) , pp. 2233-2245 (1994) ; P. Ganal , P. Moreau ,G. Ouvrard , M. Sidorov , M. McKelvy and W. Glaunsinger , “Structural Investigation of Mercury-Intercalated Titanium Disulfide . 1. The Crystal Structure of Hg1.24TiS2”, Chem. Mater. 7 (6) , pp. 1132-1139 (1995) ; idem. , “Structural Investigation of Mercury-Intercalated Titanium Disulfide . 2. HRTEM of HgxTiS2”, Chem. Mater. 7 (6) , pp. 1140-1152 (1995) ; P. Moreau , P. Ganal , S. Lemaux , G. Ouvrard , and M. McKelvey , “Mercury Sublattice Melting Transition in the Misfit Intercalation Compound Hg1.24TiS2”, J. Phys. Chem. Solids 57 (6-8) , pp. 1129-1132 (1996) ; M.V. Sidorov , M.J. McKelvey , J.M. Cowley , and W.M. Glaunsinger , “Novel Guest-Layer Behavior of Mercury Titanium Disulfide Intercalates”, Chem. Mater. 10 (11) , pp. 3290-3293 (1998) . See also : P. Moreau , G. Ouvrard , P. Gressier , P. Ganal , and J. Rouxel , “Electronic Structures and Charge Transfer in Lithium and Mercury Intercalated Titanium Disulfides”, J. Phys. Chem. Solids 57 (6-8) , pp. 1117-1122 (1996) .

ambient conductivity : L.E. Conroy and K.C. Park , Electrical Properties of the Group IV Disulfides TiS2 , ZrS2 , HfS2 , and SnS2, Inorg. Chem. 7 (3) , pp. 459-463 (1968) ; TiS2 , Fig. 2 , p. 461 .

Di Salvo and co-workers : F.J. Di Salvo et al. , “Metal Intercalation Compounds of TaS2 : Preparation and Properties, J. Chem. Phys. 59 (4) , pp. 1922-1929 (1973) ; A.C. Gossard , F.J. Di Salvo , and H. Yasuoka , “Conduction-band Formation in Metal Layers Intercalated in TaS2 : Nuclear Resonance of Sn , Hg , and Pb in TaS2, Phys. Rev. B 19 (10) , pp. 3965-3968 (1974) .

almost as strong as cyanide anion : J. March , Advanced Organic Chemistry , Reactions , Mechanisms , and Structure , 4th edition , John Wiley , New York , 1992 ; Table 10.9 , “Nucleophilicities of Some Common Reagents”, p. 351 . Sulfhydride anion , HS, is as strong a nucleophile as cyanide and iodide anions . Sulfide anion is probably comparable in nucleophilic strength .

changes to a disordered : anon. , “Exp. 3 : Thermochromism in the Ionic Conductor , Cu2HgI4, Dept. of Chemistry , University of Massachusetts , Amherst (MA) , 8 pp. ; undated [PDF , 223 KB] .

Wells : A.F. Wells , Structural Inorganic Chemistry , 3rd edition , Clarendon Press , Oxford (UK) , 1962 ; SnS is mentioned on p. 527 , and see Fig. 50(d) , p. 137 .

black , solid “premix : “The union of mercury and sulfur is effected in the most simple way of all by merely rubbing the two substances together . The product is a black , amorphous compound , quite unlike the bright scarlet crystalline mineral cinnabar” : W.E. Henderson and W.C. Fernelius , A Course in Inorganic Preparations , McGraw-Hill , New York , 1935 ; p. 58 . This useful inorganic chemistry synthesis compendium can be downloaded for free from the library resources web page [DJVU , 1396 KB ; a suitable DjVu reader for your computer can be downloaded for free from . The WinDjView reader v. 1.0.3 for older FAT 32 Windows OS can be downloaded for free from FileHorse] . The Merck Index (8th edition , 1968 , p. 661) states that this black HgS ......“usually contains about 60% HgS , the remainder being free sulfur [in the commercial product] ”. Pure black crystalline HgS has the cubic zinc blende structure .

Donohue : P.C. Donohue , “The Synthesis , Structure , and Superconducting Properties of New High-Pressure Forms of Tin Phosphide”, Inorg. Chem. 9 (2) , pp. 335-337 (1970) .

promotion energies : also called hybridization energies ; the energy required to promote native orbitals and electrons from their ground state (in degenerate atoms) to the excited state where they can blend together into the hybrid orbitals with their combined electrons : L. Pauling , The Nature of the Chemical Bond and the Structure of Molecules and Crystals , 3rd ed. , Cornell University Press , Ithaca (NY) , 1960 ; pp. 118-120 .

detonate like gunpowder : J.R. Partington , A Textbook of Inorganic Chemistry , 6th edition , Macmillan , London (UK) , 1957 ; p. 783 ;  see also C.A. Jacobson and C.A. Hampel , Encyclopedia of Chemical Reactions , Vol. 8 , Reinhold Publishing , New York , 1959 ; p. 183 . The highly exothermic zinc + sulfur reaction is vividly portrayed in the following YouTube videos : 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 reaction of aluminum and sulfur : For example , see these YouTube videos : [MP4 , 3959 KB , runtime 0:55] ; [MP4 , 1331 KB , runtime 0:23] ; [MP4 , 5826 KB , runtime 1:06] . Two different laboratory preparations of aluminum sulfide (both quite safe) are described by : G. Brauer (ed.) , Handbook of Preparative Inorganic Chemistry , Vol. 1 , 2nd edition , Academic Press , New York , 1963 ; pp. 823-824 . This exhaustive inorganic chemistry synthesis compendium (Vols. 1 & 2 combined) can be downloaded for free from the library resources web page [PDF , 19,090 KB] .

solid state metathesis reaction : J.B. Wiley and R.B. Kaner , “Rapid Solid-State Precursor Synthesis of Materials”, Science 255 (5048) , pp. 1093-1097 (1992) ; R.E. Treece , G.S. Macala , and R.B. Kaner , “Rapid Synthesis of GaP and GaAs from Solid-State Precursors”, Chem. Mater. 4 (1) , pp. 9-11 (1992) ; R.B. Kaner et al. , “Rapid Solid-State Methathesis of Refractory Materials”, U.S. Patent 5110768 , 8 pp. (May 5 , 1992) [PDF , 368 KB] ; R.B. Kaner , C.H. Wallace , and T.K. Reynolds , “Instantaneous Synthesis of Refractory Nitrides from Solid Precursors”, U.S. Patent 6096282 , 11 pp. (August 1 , 2000) [PDF , 127 KB] ; R.B. Kaner and C.H. Wallace , “Process for Rapid Solid-State Formation of Refractory Nitrides”, U.S. Patent 6120748 , 8 pp. (September 19 , 2000) [PDF , 89 KB] . Note : these three patent files can be opened only with Adobe Acrobat Reader v. 6 or later .

aluminum(0) metal and aluminum(III) oxide : M. Hoch and H.L. Johnston , “Formation , Stability and Crystal Structure of the Solid Aluminum Suboxides : Al2O and AlO”, J. Amer. Chem. Soc. 76 (9) , pp. 2560-2561 (1954) ; C.N. Cochran , “Aluminum Suboxide Formed in Reaction of Aluminum with Alumina”, J. Amer. Chem. Soc. 77 (8) , pp. 2190-2191 (1955) . See also T. Forland et al. , “Measurements of Phase Equilibria in the Aluminum – Aluminum Sulfide System”, Acta. Chem. Scand. , Series A28 (2) , pp. 226-228 (1974) [PDF , 375 KB ; DJVU , 109 KB] . The compound AlS has a narrow window of stability between 1010 C and its m.p. of 1060 C .

H. Lei and co-workers : H. Lei , K. Wang , M. Abeykoon , E.S. Bozin , and C. Petrovic , “New Layered Oxysulfide SrFBiS2”, (August 15th, 2012) [PDF , 1622 KB] .


[ Index Page ] [ Contact ]