Mol. CrySf. Liq. Cryst 19850026-S9411851l184-03i3/$10' Vol. 118. pp. 323+326© 1985G d .00/0Prj ,or on and Breach S'moo In the United S ,Clence Publishers, Inc. and OPA Ltd.

rates of America

THEORETICAL IOF SOLITON S NVESTIGATION OF A NEW CLASSPOLYARENEM-EUTPHPORTINGCONJUGATED POLYMERS'IDES .

J.L. BREDAS a •~ILBEY.. ,. D.S. BOUDREAUX,+ R.R. CHANCE! and R.

Lab. Chimie The . .de la Paix, B-5000 onque Apphqute, Fac. Universitaires Notre-Dame+ All' d Namur (Belgium)

ie Corporatio C!~rsey 07960 n, orporate Research Center, Morristown, New

Department of Ch .Cambridge M ernistry, Massachusetts Institute of Technology,, assachusetts 02139

Abstract Th .new class of o:ore.tlcal predictions for the electronic properties 01 apolymers dgamc polymers? polyarenemethides, are discussed. These

are esi dto that in t gne to possess a degenerate ground state analogouSby the unit rans:polyacetylene. The parent polymer backbone is givenented for th cell. (_Ph~nYI_CH=qUinoid=CH-). Calculations are pres-and IIdl' .e geometnc and electronic structures of the "undimerized"

merized" hai .polaron e . . cams, as well as those corresponding to sobton andwith re xcitauons in the t1dimerized" chain Significant differences

spect t I .a po yacetylene are found.

To dats e, trans-p I

h

y'te"" in p o. yacetylene (PA) is unique among conducting polymeravo ossessmg d' .P .lng exactly th a egenerate ground state, i.e. tWOgeometriC structureselerl, disto' e same total energy. This degeneracy is a consequence of apeculiarPh/tlOn and leads to possible nonlinear soliton excitatio

n,l.2 Many

of soliton dnfomena that occur in trans-PA can be explained by the presence:bservation "rects on the chains. One of the most exciting examples is theoped PA ' f 0hrdoping levels between -0.2 and 7 mol% in AsF

S- and )<3

a

4-lUg , 0 igh d 'b'I'tgesting th con uctivity without significant Pauli susceptl 1 1 y, 'tt at· I I ' than'Port pro spin ess charged solitons may play an important ro e In e

I perttes. .Who n this p hid. ICbis as aper, we describe a new polymer system, polyareoemet 1 e,11also pos yet unsynthesized. polyarenemethide (PAM) is interesting in thatPbe sesses d ' id 'og-'I nyl ring aIt a egenerate ground-state structure where the qutn

Ol

n dternation' ernation is the equivalent of the double hon

d- siogle bon

In trans- PA:I323

324 1. L. BREDAS et at.

=CH+o-CH=(_)= CHt-, JX

To explore the consequences of ground-state degeneracy in PAM, we haveperformed a series of calculations aimed at determining the energetics ofdefect formation and the atomic and charge distortions associated with them,We report here mainly on results obtained at the level of simple HUckeltheory with a bond compressibility, i.e. the quantum-chemical equivalent. ofthe Su-Schrieffer-Heeger Harniltonian.! This allows an easy comparison WIththe results previously obtained for tr ans-PA in the same theoreticalframework. 1,5 The Hackel parameters we use are the same as those weoptimized in the case of polyparaphenylene.P MNDO semiempirical Hartree-Fock geometry optimizations and Valence Effective Hamiltonian (VEH)band structure calculations have also been completed; results of these calcu-lations can be found elsewhere.P

We first investigate the geometric and electronic structures of PAMwithout defects in the undimerized (metallic) and dimerized (Peierls distort-ed) states. We then describe calculations on polarons and solitons.

CHAINS WITHOUT DEFECTS

The Huckel geometry optimization of the "undimerized" PAM chain (allring-connecting bonds of equal length) is shown below and is in very goodagreement with the MNDO optimized geometry:

+O=f Krl, ,--,- --\-. _ ----- --t: _ ---- =--1.40 A ' 1 '1 0 J

c 1.47 A1.36A

It' . d linkedIS Important to note that all rings adopt a quinoid structure an are .by allyl-like fragments. This can be understond by the fact that if the rIngsh d . t qUItea aromattc character. the two bonds between them would turn oUlong, thereby inhibiting electroo de localization along the chains. ._

Thi . . Is dlStorIS geometnc structure is unstable with respect to a Peler. dtion to a "dirnertzed" state which leads to a doubling of the unit~cell SIzean

. id b . b ve Thea qumoi - enzenoid alternation as shown in the first graph a 0 . hit bill . 't T 15sal izatton energy of the dimerized structure is 005 eV per C7 urn ". I

b d vwi . lTh PelerScan e compare WIth the 0.015 eV stabilization in trans-PA. e Vdistortion opens up a 1.052 eV gap at the Fermi level, which is 0.35 eHsmaller than in PA. The width of the highest occupied band is 1.25 eV. VE

POLY ARENEMETHIDES 325

calculationsyield . '1 . .potentialis ab :I1m ar 6r~sults and indicate that the solid-state ionizationout .2 eV, i.e. -0.5 eV smailer than in PA.

POLARONS AND SOLITONS

Uponremoval (or add' .lationslead t h ItIO':) of an electron, the geometry optimization calcu-locallattice di: t e. formatton a polaron, i.e. a radical-ion associated with a

tstortion:

50<50 ,

purely 80 % purely

quinoid~ qui no id ,,'ooid qutnoto qui no'id

-' ~ ~ ~ ~

==O=-OrTOT-tO-i-tO,td-=O=143 J l 1.38 AUI.40A Ul.41 ~8 J l 1.43;

Thepolaron defthe figure)' ect extends mainly over 5 rings (within the dashed lines inringsacqui In which the quinoid rings remain unaffected and the benzenoidpolaronis sh a strong quinoid character. The charge (spin) density withio the

s own below:

A/.DI8 0.017 0.020 0.0260= : I I / :

,aI. ,- =1~: 1-o-1rD--o--a-=o=0.013 0.011 I

0.023 0.061 0.033

%~u=· .that the 1 IC charges (spin densities) larger than 0.01 e are gIven. Notecenter of a~gest charges (spin densities) are found one ring away from thedrawnah t e. polaron. The total charge (spin deDsitY) for the seven rings

ave IS 0.79 eelectro~he presence ~f a polaron leads to the appearance of two localizedeVon el~blevels in the gap, 0.154 eV away from the band edges, i.e. 0.372agreemI er .Slde of the Fermi level (center of the gap). This is in complete

f' ent With th . . h .leld-theor '. e value that would be predicted WlthlO. t .e contlOU

Um

abnut00/ limit as derived for trans-PA'? The polaron blOdlOg energy IS

and~O.05 eV: which is to be compared with _0.03 eV In polyparaphenylene5

W eV 10 trans_PA.5,7<aleularhen .. second electron is removed from (or added to) the chaiD, ourt IOns IOd' V . h to the f reate that two polarons are uDstable by 0.35 e Wit respec, ormaf . .IS 0 329 ron of two isolated charged solitoDS The soh ton creaUOD eDergy

0' ev- th . . t 7 ) Id b.335 eV ' e continuum limit value (equal to the gap over" ' wo

u. e

areillu t' The geometric structure and charge deDsity (in e) of the sohtoDs rated below:

326 1. L. BREDAS et al.

purely 40 %

quinoid quinoid

, {), {), ,

=O=-b--Ot-TOt-,D-=O=J 1.40,lJ 1.36,J L1.41~

_ i _ 0.03\ ~f" 0.01' l: __......=0=-0--0---0-1--0-70-), 0.058 '0.011 0,020

The soliton extends over .....4 rings and the charge density on the middl~ si~eiszero, in marked contrast with the PA case. The total charge for t?e .SIX flnfsdrawn above is 0.90 e. In the neutral soliton. the same charactenstics app yfor the spin density.

In summary, (i) polyarenemethide should be easily both p-do~e~(ionization potential around 4 eV) and n-doped (electron aff inity aro~neV); (Ii) on doping, first polarons should be formed and then recombme ~oform charged solitons, since the creation energy of two solitons is low~r hY0.35 eV than t?at of two polarons; (iii) due to the small.e~ bandWl~\:(HOMO bandwidth IS -1.25 eV), the maximum conductivit ies shaullower than in polyacetylene (where the HOMO bandwidth is -5 eV).

REFERENCES

3. Chercheur qualifie of the Belgian National Fund for Scientific Research(FNRS).1. W.P. Su, J.R. Schrieffer, and A.I. Heeger, Phys. Rev. !! 22,2209 (1980).2. M.I. Rice, Phys. Lett. A 71, 152 (1979).

---- T Wefner,3. D. Moses, A. Denenstein, J. Chen, A.I. Heeger, P. McAndrew, . 0

A.G. McDiarmid, and Y.W. Park, Phys. Rev. B 25, 7652 (1982). B 294. T.C. Chung, F. Moraes, J.D. Flood, and-A.J. Heeger, Phys. ~ __ '2341 (1984). 195. J.L. Bredas, R.R. Chance, and R. Silbey, Mol. Cryst. Liq. Crys!. 72, 3(1981); Phys. Rev.!! 26, 5643 (1982). -- r J L.6. D.S. Boudrea~x, R.R. Chance, R.L. Elsenbaumer, J.E. ~romme , Am.Bredas, and R. Silbey, Phys. Rev. B, submitted for publtcatton: ~ ~Phys. Soc. 29, 316 (1984). - - t 77.7. A.R. Bishop, D.K. Campbell, and K. Fesser, Mol. Crys!. Liq, ~_253 (1981). -