Electronically-active Triangulene Compounds

 

In this web page I'll discuss the electronic structure and proposed syntheses of new compounds based on the triangulene molecule :

Triangulene itself , while thought to be electronically aromatic (it has 22 p electrons in benzenoid rings) , apparently is too reactive to be isolated . Attempts by Clar and Stewart in 1953 to synthesize triangulene proved futile , as the neutral free radical very rapidly polymerized under their experimental conditions (the references are below , at the end of the page) . There are three non-equivalent dihydrotriangulene isomers , which I will call A , B, and C :

Surprisingly , even these (or at least the A isomer , which Clar and Stewart investigated) were also extremely reactive , and couldn't be isolated pure . For example , the A isomer underwent a very fast oxidation at room temperature :

Methylolation of triphenylmethane with formaldehyde could install methylene bridges across the phenyl groups , thus providing a simple , straightforward entry into the triangulene system :

In the above scheme , tetrahydrotriangulene is subjected to hydride abstraction by the trityl cation (the hexafluorophosphate anion is an arbitrary selection) to provide the corresponding triangulenium cation .

An alternate source of the methylene bridges could be dichloromethane , installed by a Friedel-Crafts alkylation . Diphenylmethane was prepared from benzene and dichloromethane by such a reaction :

Even more interesting would be the sulfur analogues of triangulene sketched below :

In numbering the atoms of these molecules , I have given priority to the sulfur atoms , which are the largest and heaviest components of the structures . This numbering of the triangulene molecule is different from that of both Laursen [see in the sketch above of the dihydrotriangulenes A , B, and C] and Clar and Stewart . Also , I have given the central triangulene atom , whether of carbon or nitrogen , the number 13 , as I have no particular numerological bias against this designation .

The aza analogue is reminiscent of methylene blue , the familiar deep blue cationic dye :

Suppose these compounds could be synthesized , then subjected to hydride abstraction (hydrotriangulene) or one-electron oxidation (azatriangulene) . Aromatization could then theoretically occur in the molecules , and in the case of azatriangulene might produce startling results . The aza compound could become a molecular metal :

In the latter case the nitrogen atom's radical cation is quite energetic . It can reduce its energy by becoming delocalized over the molecule by resonance . In order for this to occur the sulfur atom must rehybridize from tetrahedral sp3 to trigonal planar sp2 (as in the thiophene molecule , for example) . However , there will be seven p electrons in the ring . If one of them can be promoted to an external ("frontier") orbital over the sulfur atom , then the ring can be energetically stabilized by aromatization of the remaining six pi electrons .

The seventh pi electron could be promoted into an antibonding MO over the molecule . Another more interesting possibility would be for it to enter the empty 4s AO over the sulfur atom . This may actually be happening in the molecular metals containing electronically active sulfur atoms , such as TTF-TCNQ and in the metallic inorganic polymer , poly(sulfur nitride) , (SN)x . When the 4s AOs with their single electrons overlap continuously in the crystalline solid , a sigma XO (crystal orbital , metallic bond , conduction band) will be formed . This metallic bond will also provide some stabilizing energy to the system as all chemical bonds do , of course . The overall stabilization energy provided by the aromatization , the metallic bond , and the van der Waals bonds in the crystalline solid may be sufficient to overcome the promotion energy required to transfer the seventh pi electron from the 3p orbital (valence band) up into the 4s frontier orbital (conduction band) . If this transfer can actually occur , then the azatriangulenium radical cation salt may be a molecular metal .

Synthesis of the sulfur analogues of triangulene might be readily accomplished by the reaction of elementary sulfur with triphenylmethane or triphenylamine accompanied by a Lewis acid catalyst , as in the Friedel-Crafts reaction . Anhydrous aluminum chloride was used for this purpose in the preparation of phenoxthin from diphenyl ether :

Preparation of 1,5,9-trithiatriangulene and its carbenium salt could proceed analogously :

As mentioned above , in the carbenium salt the rings with the sulfur atoms are expected to be aromatic .This proved to be difficult to illustrate with my chemistry graphics software . However , the 1,5,9-triaza (TATA) analogues , sketched below in the References and Notes section , are similarly predicted to be fully aromatic , and I was able to show this in the drawing .

I have suggested the use of carbon disulfide as the solvent in the above synthesis . It has long been a popular solvent in the Friedel-Crafts reaction , especially when solid reactants are involved . In the phenoxthin preparation the inexpensive diphenyl ether was used in a large excess as the reaction solvent ; however this option is unavailable here . Carbon disulfide might have a useful advantage in these proposed syntheses , as it is the best known solvent for elementary sulfur ("flowers of sulfur" , a fine powder , is generally used in organic reactions) , dissolving 230 g per litre at 0 ºC (the Merck Index says 1g/2ml , or 500g/L , but that might be the hot solvent) . Carbon disulfide has several disadvantages : its relatively low boiling point (46 ºC) , foul smell and toxicity (fume hood !) , and high flammability , similar to that of ethyl ether and petroleum ether . It could be used as the solvent in these sulfurization reactions , but carefully ; if heating the reaction mixture is required , it should be done on a steam bath , with swirling by hand : no electrical heating or stirring ! Also , the highly toxic by-product hydrogen sulfide gas is evolved in the preparation . It can be vented with efficient aspiration in a fume hood , or bubbled through wash bottles to be absorbed by (possibly) sodium hypochlorite solution (Javelle water) - which oxidizes it to sulfur - and by ferric chloride solution , whose iron(III) cations bond strongly to sulfur and precipitates it as iron sulfides .

The aza derivative might be synthesized in a similar manner :

Noting in my chemicals catalog (Aldrich Catalog Handbook of Fine Chemicals) that the starting material , triphenylamine , is somewhat expensive , I included in the synthesis scheme a possible preparation of it from the inexpensive chlorobenzene , via a nucleophilic displacement by nitride anion . A very polar organic solvent such as propylene carbonate would have to be used in the reaction , to stabilize the ionic intermediate in the otherwise unreactive chlorobenzene . I have discussed these sorts of nucleophilic displacements in highly polar solvents in another web page . Such a simplified preparation of triphenylamine would be an interesting project by itself .

Treatment of the trithia-aza-triangulene product 2 with a suitable one-electron oxidizing agent should remove an electron from the central nitrogen atom , converting the compound into a radical cation aminium salt . This functionality is found in the interesting Wurster Blue salts , such as :

Another fascinating example of a Wurster Blue , tris(4-bromophenyl)aminium hexachloroantimonate , is the subject of another web page . As pointed out above , if the radical cation is energetically capable of delocalization , the "seventh" pi electrons from the three sulfur atoms will be promoted into a frontier orbital over them . If this is the sulfur 4s AO , the aminium salt should display metallic properties as a molecular metal :

The sulfur 4s AOs should overlap continuously throughout the crystalline solid to produce the sigma XO (metallic bond) in it . The promoted 4s electrons will be assigned to a vast multitude of energy levels by the Fermi-Dirac distribution , such that most of them (typically 99%) are paired off in energy space , with about 1% of them unpaired and above the Fermi level . Suppose we were to treat the trithia-aza salt with a second equivalent of one-electron oxidizer ; the strong oxidizing agent nitrosonium hexafluorophosphate is suggested for this purpose . The oxidizer would remove one of the three promoted 4s electrons , leaving two in the sigma XO over the molecule , plus the positive "hole" . These two 4s electrons would be delocalized over the three sulfur atoms in an intramolecular mixed-valence resonance . They would form an aromatic electron pair (4n + 2 , n = 0) in the sigma XO , like the three resonant pairs of p electrons in the benzene molecule . In benzene the pi MO electron pairs are confined to the molecular ring , so it's an electrical insulator . In the trithia-aza salt's sigma XO the electron pairs will be mobile , and should be able to move through the crystal-wide XO under an applied potential difference . That is , they will constitute the electrical current in it . If they can migrate from molecule to molecule intact , and at room temperature and one atmosphere pressure , the compound will be an ambient superconductor . The electron pairs will be functional Cooper pairs (I'm assuming that all the aromatic pairs of electrons in the sulfurs' sigma XO will be above the Fermi level , which could occur at the interface between the lower energy level aromatic pi MO and the higher energy level XO) :

Note that this sort of reasoning has been applied in another example of an organosulfur compound originally designed as a molecular metal but which might also be an ambient superconductor , 1,3,5-trithiabenzenium hexafluorophosphate ; this material has been discussed in another web page . However , the trithia-aza-triangulenium salt should be more readily accessible than the former compound , making it of greater interest to solid state chemistry and physics researchers .

 

References and Notes

 

Clar : E. Clar and D.G. Stewart , “Aromatic Hydrocarbons . LXV . Triangulene Derivatives”, J. Amer. Chem. Soc. 75 (11) , pp. 2667-2672 (1953) .

Methylolation : Generally , the insertion of a methylene bridge between two reagents A-H and B-H having reasonably labile hydrogen atoms , using formaldehyde as the methylene source , with expulsion of a water molecule as the reaction by-product :

A–H   +   HCHO   +   H–B   ------------------>    A–CH2–B   +   H2O

Many reagents can be used as the A-H and B-H components . The most famous (and useful) type of methylolation is the Mannich reaction (1912) , an example of which is :

CH3–CO–CH3   +   HCHO   +   (CH3)2NH   --------------->

                                                    CH3–CO–CH2–CH2–N(CH3)2   +   H2O

I have discussed methylolation , with more examples , in another web page .

trityl cation : The trityl (triphenylcarbenium) cation has been used as a hydride extraction reagent in cases where aromatic or otherwise energetically stabilized reaction products are formed . See for example : W. Bonthrone and D.H. Reid , “Hydride Ions in Organic Reactions . Part I . Dehydrogenation by Triphenylmethyl Perchlorate”, J. Chem. Soc. , pp. 2773-2779 (1959) . Two interesting examples from their research are sketched below :

Trityl carbocation reagents are salts , Ph3C+ X- , where “X” is an inert spectator anion , such as ClO4- , BF4- , PF6- , AsF6- , SbF6- and SnCl5- . They are commercially available (Aldrich , Alfa-Aesar) , but are rather expensive . One method of preparing them is by combining trityl alcohol [triphenylcarbinol] with the corresponding mineral acid :

 Ph3C–OH   +   HBF4   ----------------->   Ph3C+ BF4-   +   H2O 

See for example : H.J. Dauben Jr. , L.R. Honnen , and K.M. Harmon , “Improved Preparation of Triphenylmethyl Perchlorate and Fluoroborate for Use in Hydride Ion Exchange Reactions”, J. Org. Chem. 25 (8) , pp. 1442-1445 (1960) .

Friedel-Crafts : C.C. Price , “The Alkylation of Aromatic Compounds by the Friedel-Crafts Method” , Organic Reactions , Vol. 3 , Ch. 1 , pp. 1-82 , R. Adams (ed.) , John Wiley , New York , 1946 ; see especially p. 12 ; G.A. Olah and D. Meidar , “Friedel-Crafts Reactions”, Kirk-Othmer Encyclopedia of Chemical Technology , Vol. 11 , pp. 251-268 , M. Grayson and D. Eckroth (eds.) , John Wiley , New York , 1980 ; C.A. Thomas , Anhydrous Aluminum Chloride in Organic Chemistry , Reinhold Publishing , New York , 1941 ; pp. 108-109 ; Wikipedia : “Friedel-Crafts reaction” at : http://en.wikipedia.org/wiki/Friedel-Crafts_reaction .

Laursen : B.W. Laursen , “Triangulenium Salts”, Ph.D Thesis , Copenhagen , Denmark , June 2001 . This comprehensive study of triangulene chemistry can be downloaded without charge from the Web (PDF document , 2372 KB) at : http://www.risoe.dk/rispubl/POL/polpdf/ris-r-1275.pdf . Laursen and Krebs have synthesized other interesting triangulene derivatives such as "TATA" , triazatriangulenium salts :

Note the fully aromatic electronic structure of the TATA compounds , similar to what was predicted above for the trithiatriangulenium and aza derivatives .

TTF-TCNQ : E.M. Engler , “Organic Metals”, Chemtech 6 (4) , pp. 274-279 (April , 1976) .

poly(sulfur nitride) : A.G. MacDiarmid et al. , “Synthesis and Selected Properties of  Polymeric Sulfur Nitride (Polythiazyl) , (SN)x”, Ch. 6 , pp. 63-72 in Inorganic Compounds with Unusual Properties , R.B. King (ed.) , Adv. Chem. Series 150 , American Chemical Society , Washington , D.C. , 1976 ; M.M. Labes , P. Love , and L.F. Nichols , “Polysulfur Nitride – A Metallic , Superconducting Polymer”, Chem. Rev. 79 (1) , pp. 1-15 (1979) .

phenoxthin : C.M. Suter and C.E. Maxwell , “Phenoxthin”, Org. Synth. Coll. Vol. 2 , pp. 485-486 (1943) . The procedure can be downloaded without charge from the Web (PDF document , 111 KB) at : http://www.orgsyn.org/orgsyn/pdfs/CV2P0485.pdf .

popular solvent : For example , see in Price's review (cited above) of the Friedel-Crafts reaction , Table VII , “Alkylation of Napthalene” , pp. 53-55 , and Table XV , “Alkylation of Heterocyclic Aromatic Compounds”, pp. 76-77 , where carbon disulfide is widely employed as the reaction solvent .

triphenylamine : This reagent has been prepared by the alkylation of diphenylamine by iodobenzene with catalysis by a small amount of copper or copper bronze powder : F.D. Hager , “Triphenylamine”, Org. Synth. Coll. Vol. 1 , pp. 544-547 (1941) . The procedure can be downloaded without charge from the Web (PDF document , 117 KB) at : http://www.orgsyn.org/orgsyn/pdfs/CV1P0544.pdf .

Wurster Blue : See the discussion , photo , and and video demonstration of the formation of a Wurster Blue compound in solution , from Peter Keusch's web page , “Wurster's Blue”, at http://www.uni-regensburg.de/Fakultaeten/nat_Fak_IV/Organische_Chemie/Didaktik/Keusch/D-Wurster-e.htm

Fermi level : A.R. Mackintosh , “The Fermi Surface of Metals”, Scientific American , 209 (1) , pp. 110-120 (July , 1963) .

nitrosonium : W.J. Plieth , “Nitrogen”, Ch. 5 , pp. 321-479 in Encyclopedia of Electrochemistry of the Elements , Vol. 8 , A.J. Bard (ed.) , Marcel Dekker , New York , 1978 ; p. 325 . The redox half-reaction involved is :

NO+   +   e-      --------------->    NO (g)      E0red  =  1.45 V

Several of the nitrosonium salts are commercially available , eg. from Aldrich Chemicals and Alfa-Aesar .

migrate : J.H. Perlstein , “Organic Metals – The Intermolecular Migration of Aromaticity”, Angew. Chem. Internat. Ed. Engl. 16 (8) , pp. 519-534 (1977) ; see especially Fig. 4 , p. 526 .

 

 

[ Index Page ] [ Contact ]