This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

Bonding in Cyclophosphazenes: Quantum and Experimental Support for DEWAR'S Island Model *

J . - P . F A U C H E R * * , J . - F . L A B A R R E * * , a n d R . A . S H A W * * *

** Laboratoire de Chimie de Coordination du CNRS BP 4142, 31030 Toulouse Cedex, France *** Birkbeck College, Department of Chemistry, Malet Street, London WC1E 7HX, U. K.

(Z. Naturforsch. 31b. 677-679 [1976]; received July 27, 1975)

Transannular Bonding, Inorganic Ring Systems, CNDO/2 Calculation, Cyclophosphazenes, Dewar's Island Theory

Quantum mechanical calculations for a number of cyclophosphazenes reveal that transannular bonding and antibonding interactions contribute considerably to the stability and conformation of these PN ring systems. Charge distributions are only in agreement with the island theory; the latter is confirmed by Faraday effect measurements.

The electronic structure of some symmetrically substituted cyclophosphazenes was investigated prite recently for the first t ime1 within the frame-vork of the now classical C N D O / 2 approximation2

n order to differentiate between the two drastically •pposite assumptions which have been developed iround 1960 about the nature of the electronic lelocalization in these inorganic rings.

The net charges and the W I B E R G indices3 cal-culated for (NPF2)3 , (NPC12)3 and (NPF2)4 are ound in Fig. 1 and these definitely support D E W A R ' S

hree-center island model4 , rather than C R A I G and ^ D D O C K ' S 'benzene-like' one5 .

Although the W I B E R G indices of endocyclic P - N

>onds are large (thus favouring an important de-ocalization within the ring) the lack of electron lensity on the phosphorus atoms is so great that IO ring current can exist, the derealization remain-tig ' localized' within the PNP units6. Moreover, the dgh electron density on the nitrogen atoms corre-ates with the remarkable basicity properties of hese donor sites.

However, a question arose at this stage: H o w i it possible to explain on the basis of D E W T A R ' S

aodel the high stability of phosphazene rings ? 'he answer was provided by quantum chemistry:

Paper, presented at the 1. International Symposium on Inorganic Heterocycles, Besan§on (France), June 16-19, 1975.

Requests for reprints should be sent to J.-P. FAU-HER, Laboratoire de Chimie de Coordination du NRS BP 4142, F-31030 Toulouse Cedex, France.

F F Cl Cl

Fig. 1

P E R K I N S 7 and ourselves8 have shown that in (NPX2)3 trimers there exists a strong transannular bonding interaction between endocyclic P atoms, the energy of which is about 1 0 % of the P - N bond energy (Fig. 2). For (NPX2)4 tetramers, the nature o f these interactions is more complicated: the stability and the preferred conformations of such rings depend upon the balance of two terms (i) a bonding component TADJ. identical to the one mentioned above for ( N P X 2 ) 3 (ii) and an antibonding term T 0 P P . (Fig. 2)8 .

678 J. P. FAUCHER ET AL. • BONDING IN CYCLOPHOSPHAZENES

!

V Fig. 2

Such through-space interactions exist also in other inorganic ring systems. Quantum chemistry shows for example that the tub form of N4S4 is due to important (S. . . S) bonding terms9, the order of magnitude of which - with respect to the S - N bond energy - is still larger than in P - N rings.

Similarly it was shown that the preferred chair conformation of (Me2NAlH2)310 is due to consider-able transannular (A1...A1) bonding interactions amounting to 50% of the Al-N bond energy.

Consequently, it must be emphasized that caution must be used with the rough classical description of molecules based on the 'chemical bond' pattern. As soon as a molecule contains at least two atoms of the second row of the periodic table11, the energy (and all the monoelectronic quantities) are spread out over the whole molecule and ob-viously the concept of chemical bond - based on the principle of topological proximity - is unable by itself to provide a complete description of chemical and physical behaviour.

These reflections suggest that chemists ought to keep an open mind on the concept of chemical bonding.

The results which we have presented so far in order to support D E W A E ' S model are theoretical ones and definite experimental proof for the validity of this model was needed.

Such a proof was provided by means of Faraday effect measurements on (NPX2)3,4,5 compounds: For a given X the molecular rotations throughout the corresponding series are exactly - with respect to the accuracy of the method - in the 3 : 4 : 5 ratio and this result is only consistent with the existence of individual PNP entities12. If, indeed, the derealiza-tion in cyclophosphazenes would have been of the 'benzene-like' type, the molecular magnetic rotation would have to increase exponentially with the size of the ring6.

Thus, the validity of D E W A R ' S model for cyclo-phosphazenes is theoretically and experimentally clearly demonstrated. Encouraged by the success of these preliminary studies, we extended the CNDO/2 method to obtain information on the electronic struc-ture in the ground state of many cyclophosphazenes in order to provide the experimentalists with data needed to interpret their results more fully.

Two recent investigations in this field will be pointed our briefly in this contribution:

The electron distributions of the three isomeric N3PsCl3(NMe2)3 allowed us13 to rationalize the set of physicochemical results which had been obtained on them ( W NMR, 35C1 NQR, pK'a data, leis.-effect', . . .).

The protonation site of cyclophosphazenes was investigated for the highly basic N3P3(NMe2)e 14

Quantum chemical calculations show that, in full agreement with experiment, protonation takes placc on the endocyclic nitrogen sites and not on the NMes groups. More importantly, the CNDO/2 approach predicts a lengthening of the P - N bonds of the rin£ and a simultaneous shortening of exocyclic P-N bonds when protonation occurs. These predictions were nicely illustrated by a recent crystallographic study performed by A L L C O C K 1 5 on the N3P3(NMe2)< protonated species.

Even more important is the fact that protonatioi induces in the N3Ps(NMe2)6 molecule a drastic in version of the relative stability of the endo ant exocyclic P - N bonds: in other words, the regenera tive (i.e. the stable) character of the phosphazent ring vanishes in the protonated form in favour o an enhancement of the stability of P-NMe2 linkages

Thus we assume that the chemical behaviour o protonated cyclophosphazenes will be very differen from that of the free bases.

In conclusion we believe that the concerted use o quantum chemistry and Faraday effect has providet definite proof of the validity of D E W A R ' S mode to describe the electronic structure in the groun< state of cyclophosphazenes. The 'benzene-like model must be therefore abandoned and cyclo phosphazenes must be labelled as non-strobili compounds6.

Such simultaneous theoretical and experimenta approaches will be applied to many other series o inorganic rings in order to attempt to unify th description of these rings within a generalization o D E W A R ' S model.

J. P. FAUCHER ET AL. • BONDING IN CYCLOPHOSPHAZENES 679

J . - P . F A U C H E R , J . D E V A N N E A U X , C . L E I B O V I C I , a n d J.-F. LABARRE, J. Mol. Structure 10, 439 [1971]. J. A . POPLE and G . A . SEGAL, J. Chem. Phys. 43, 136 [1965]. K. WIBERG, Tetrahedron 24, 1083 [1968]. M . J . S . D E W A R , E . A . C . L Ü C K E N , a n d M . A . W H I T E H E A D , J . C h e m . S o c . 1 9 6 0 , 2 4 2 3 . D . P . C R A I G , C h e m . I n d . 1 9 5 8 , 3 ; D . P . C R A I G a n d N. L. PADDOCK, Nature 181, 1052 [1958]; D. P. C R A I G , M . L . H E F F E R N A N , R . M A S O N , a n d N . L . PADDOCK, J. Chem. Soc. 1961, 1376. J . - F . LABARRE and F . CRASNIER, Topics in Current Chemistry 24, 33 [1971], D . R . A R M S T R O N G , G . H . L O N G M U I R , a n d P . G . PERKINS, Chem. Commun. 1972, 464. J . - P . F A U C H E R a n d J . - F . L A B A R R E , P h o s p h o r u s 3 , 265 [1974].

9 P . CASSOUX , J . - F . L A B A R R E , O . G L E M S E R , a n d W. KOCH, J. Mol. Structure 13, 405 [1972].

1 0 M . PELISSIER , J . - F . L A B A R R E , L . V . V I L K O V , A . V . G O L U B I N S K Y , a n d V . S . M A S T R Y U K O V , J . C h i m . Phys. Fr. 70, 702 [1974].

1 1 J . - F . L A B A R R E a n d C . L E I B O V I C I , I n t e r n a t . J . Quantum Chem. 6, 625 [1972]; G. ROBINET, J.-F. LABARRE, and C. LEIBOVICI , Chem. Phvs. Letters 15, 90 [1972],

1 2 J . - P . F A U C H E R , O . G L E M S E R , J . - F . L A B A R R E , a n d R. A. SHAW, C. R. Acad. Sei. 279C, 441 [1974].

1 3 J . - P . F A U C H E R , J . - F . L A B A R R E , a n d R . A . S H A W , J. Mol. Structure 25, 109 [1975].

1 4 J . - P . F A U C H E R a n d J . - F . L A B A R R E , A d v a n c e s in M o l . Relax. Interact. Processes, in press.

1 5 H . R . A L L C O C K , E . C . B I S S E L L , a n d E . T . S H A W L , Inorg. Chem. 12, 2963 [1973].