Spectral stability study and molecular modeling of fluorence based conjugated polymers 5

85
CHAPTER 5 INFLUENCE OF DONOR AND ACCEPTOR
SUBSTITUENTS ON THE ELECTRONIC CHARACTERISTICS
OF POLY(FLUORENE-PHENYLENE)



86
5-1 Introduction

Since the discovery of electroluminescence of poly(p-phenylenevinylene)
1
, a
conjugated polymer, polymer light-emitting diodes (PLEDs) have drawn special
attention because of the potential applications in developing large area and flexible
displays
2-4
. The design of light-emitting diodes (LEDs) involving an organic
conjugated system as the active layer has been the focus of a considerable amount of
theoretical and experimental studies in the past few years
1,5
. Several strategies have
been set up to improve the performances of the devices, among which derivatization of
the conjugated backbone has been found to have a significant impact
6
. Indeed, this
approach allows for optimizing the match between the frontier electronic levels of the
polymer and the Fermi energies of stable metallic electrodes, and hence for a better
balance of electron and hole injection rates, which is required for the achivement of
high efficiencies. These considerations have also led to the fabrication of efficient

multilayer LED devices including electron and/or hole transporting layers
7
.

Here, we investigate theoretically the influence of the presence of chemically
simple acceptor and/or donor groups as “side-groups” along a poly(9,9-
dihexylfluorene-1,4-phenylene) chain. The present quantum chemistry calculations
report on the changes in the geometric, electronic properties of the poly(9,9-
dihexylfluorene-1,4-phenylene) unit cell that are induced by substitution with an
acceptor such as a cyano group or a donor such as a methoxy or an amino group. We
stress that the trends derived from this study fully represent of the properties observed
at the scale of long polymer chains.




87
5-2 Theoretical methodology

The chemical structures of the polymers I-VIII are shown in Figure 5.1. The
polymers can be divided into four groups: (i) poly(9,9-dihexylfluorene-1,4-phenylene),
which serves as reference; (ii) poly(9,9-dihexylfluorene-2,5-dimethoxyl-1,4-
phenylene), a compound representing of the dialkoxy derivatives, often used in
devices; (iii) 2-amino, 5-amino and 2,5-diamino derivatives in which the amino
group(s) substitutes the phenylene group; and (iv) 2-cyano, 5-cyano and 2,5-dicyano
derivatives, with the cyano group(s) as substituent(s) on the phenylene group.
Semiempirical SCF MO calculations were performed using the AMPAC 6.51
program
8
installed on a SGI Origin 2000 workstation. The optimized geometries and

the one-electron structures of the polymer unit cells were obtained by using the Austin
Model 1 (AM1) method with MNDO as the Hamiltonian.















88

V
III
II
I
n
n
n
n
n
NH
2
NH
2
NH
2
C
6
H
13
C
6
H
13
C
6
H

13
C
6
H
13
NH
2
C
6
H
13
C
6
H
13
OCH
3
OC
H
3
C
6
H
13
C
6
H
13
C
6

H
13
C
6
H
13
VIII
VII
VI
n
n
n
CN
CN
CN
CN
C
6
H
13
C
6
H
13
C
6
H
13
C
6

H
13
C
6
H
13
C
6
H
13
IV


Figure 5.1 Chemical structures of the various PDHFP derivatives

5-3 Results and discussion

5-3-1 Geometry considerations

The calculated dihedral angles for the various unit cells are shown in Table 5.1.
We note that the dihedral angle between the plane of phenylene and fluorene for I is -
41.2 º from the AM1 calculation results. Upon substitution of the hydrogen atoms at 2
and 5 positions of the phenylene ring with two methoxy groups in II, the dihedral angle
is very similar to that of I, showing that the methoxy groups have not much effect on



89
the dihedral angle. Upon substitution of the hydrogen atom at 2 (III) or 5 position (IV)
and two hydrogen atoms at 2 and 5 positions (V) of the phenylene group with amino

group(s), the dihedral angle are - 62.0, - 41.4 and - 62.6, respectively. It seems that the
amino group at 2 position of the phenylene group affects the dihedral angle in a great
degree while the amino group at 5 position has not much effect on the dihedral angle.
That the dihedral angle of V is very similar to that of III confirms this point. The
situation is almost the same when the hydrogen atom at 2 (VI) or 5 position (VII) and
two hydrogen atoms at 2 and 5 positions (VIII) of the phenylene group are substituted
by cyano group(s).

Table 5.1 Dihedral angles between the plane of phenylene and fluorene for the
various unit cells

Oligomer Dihedral angle (º)
I - 41.2
II - 41.1
III - 62.0
IV - 41.4
V - 62.6
VI - 54.8
VII - 41.3
VIII - 55.3



5-3-2 One-electron structure

Here, we focus our attention on the way the locations of the HOMO (highest
occupied molecular orbital) level and LUMO (lowest unoccupied molecular orbital)
level are affected upon derivatization with respect to the unsubstituted 9,9-
dihexylfluorene-1,4-phenylene; the results are shown in Figure 5.2 and Table 5.2.

90










VIII
VIIVI
III
II
I
E (eV)
HOMO
LUMO
-10
-8
-6
-4
-2
0
IV

V

Figure 5.2 AM1 calculated energies of HOMO and LUMO of the various unit cells


Table 5.2 AM1 calculated shifts (in eV) of HOMO and LUMO for the various
unit cells II-VIII with respect to I

Oligomer HOMO LUMO
II + 0.07 + 0.02
III + 0.17 + 0.10
IV + 0.10 + 0.04
V + 0.63 + 0.10
VI - 0.19 - 0.20
VII - 0.21 - 0.26
VIII - 0.36 - 0.72



The band gap of II is calculated to decrease by 0.05 eV with respect to that of I.
This band gap lowering is due to asymmetric destabilizations of the HOMO and
LUMO levels with respect to I: the energy of HOMO increases by 0.07 eV and that of
LUMO increases by 0.02 eV. This destabilization effect is very clearly due to the π–
electron donating character of the methoxy groups. This behavior is further
rationalized by a detailed analysis of linear combination of atomic orbital (LCAO)



91
coefficients for the HOMO and LUMO levels; see Figure 5.3(a). The backbone

HOMO level is found to be more affected than the LUMO level upon substitution with
donor groups; the asymmetry of destabilization is characterized by a stronger
antibonding character of the C-O bond in the HOMO wave function. The absolute
value of the shift of the HOMO level is governed by the strength of the electronic
coupling between the substituting group and the backbone. The destabilization effect
has been observed experimentally in electrochemical study for poly(2,5-
dimethoxyparaphenylene vinylene)
9
.
HOMO
LUMO
(a) De
r
ivative II
(b) Derivative V
HOMO
LUMO
(c) Derivative VIII
HOMO
LUMO

Figure 5.3 Sketch of the AM1 LCAO coefficients for the HOMO and LUMO
levels in (a) derivative II, (b) derivative V and (c) derivative VIII.



92
In the case of monoamino substitution, an overall destabilization of the HOMO
and LUMO levels is expected, due to the donor nature of the substituents.
Furthermore, it is of interest to evaluate which substitution effect is stronger when the

amino group is located at 2 position or 5 position of the phenylene ring. From the
results shown in Table 5.2, we observe that the influence of substitution at 2 position is
stronger than that at 5 position. This is in agreement with the trend of the dihedral
angle change at 2 position in III and at 5 position in IV relative to I. With respect to I,
the energy of HOMO in the monoamino derivatives increases by 0.17 and 0.10 eV for
2 position and 5 position substitutions, respectively; the energy of LUMO increases by
0.10 and 0.04 eV, respectively. We have also considered the diamino substitution on
the phenylene ring in V. The results of the calculations, Figure 5.2 and Table 5.2,
indicate a strong destabilization of the HOMO level, resulting in an energy which
increases by 0.63 eV with respect to I. Relative to I, the energy of LUMO in V
increases by 0.10. The combined evolution of the HOMO and LUMO levels results in
the band gap that is shifted to the red, decreasing by 0.53 eV. This value is
significantly higher than that in dimethoxyl and monoamino derivatives discussed
above. The overall destabilization of the HOMO and LUMO levels caused by the
donor nature of the amino substituents is illustrated by the analysis of the LCAO
coefficients; see Figure 5.3(b). The situation is very similar to that of derivative II. The
larger difference between LUMO and HOMO levels induced by diamino groups in
derivative V relative to that induced by dimethoxy groups in derivative II confirms that
the decrease of band gap in derivative V relative to derivative I is larger than that in
derivative II. The calculation result for derivative V is in accordance with the
experimental result reported before.
10
In solution, the UV-visible absorption maximum



93
of poly(2,5-bis[N-methyl-N-hexylamino]phenylene vinylene) has been observed to
shift about 40 nm to the red of unsubstituted poly(paraphenylene vinylene).

In the case of monocyano substitution, the acceptor nature of the cyano
substituents gives rise to an overall stabilization of the HOMO and LUMO levels in VI
and VII. With respect to I, the energy of HOMO in the monocyano derivatives
decreases by 0.19 and 0.21 eV for 2 position and 5 position substitutions, respectively;
the energy of LUMO decreases by 0.20 and 0.26 eV, respectively. The calculation
results on the dicyano substitution in VIII shows a strong stabilization of the LUMO
level, with the energy decreasing by 0.72 eV with respect to I. Relative to I, the energy
of HOMO in VIII decreases by 0.36. Thus, the band gap is estimated to decrease by
0.36 eV that is shifted to the red. This value is much higher than that of either of the
monocyano derivatives. The overall stabilization of the HOMO and LUMO levels
caused by the acceptor nature of the cyano substituents is also illustrated by the
analysis of the LCAO coefficients; see Figure 5.3(c). The LUMO level is found to be
more affected upon substitution with acceptor groups; the asymmetry of stabilization is
rationalized by the stronger bonding character, found in the LUMO wave function, for
the bond between the carbon atom of the cyano group and the adjacent carbon atom on
the phenylene ring. The large shifts of the frontier levels in derivative VIII relative to
derivative II reflect the stronger coupling of the cyano groups to the monomer unit
with respect to the methoxy groups. Also, the larger difference between LUMO and
HOMO levels induced by dicyano groups in derivative VIII relative to that induced by
dimethoxy groups in derivative II confirms that the decrease of band gap in derivative
VIII relative to derivative I is larger than that in derivative II. The calculation result for
derivative VIII is in agreement with the dicyano substitution in poly(2,5-dicyano-1,4-



94
phenylene vinylene) experimently. The experimental band gap of MEH-PPV and M-
DCN-11 have been reported to be 2.10 and 1.92 eV, respectively
11
. M-DCN-11 is a

copolymer of MEH-PPV and poly(2,5-dicyano-1,4-phenylene vinylene), its band gap
value is intermediate (mid-way) between that of MEH-PPV and poly(2,5-dicyano-1,4-
phenylene vinylene). Therefore we can infer that the band gap of poly(2,5-dicyano-
1,4-phenylene vinylene) is less than 1.92 eV. The band gap of poly(paraphenylene
vinylene) is 2.4 eV deduced from optical absorption experiments on well-ordered
samples
12
. Thus, the band gap is decreased by more than 0.48 eV upon dicyano
substitution on poly(paraphenylene vinylene).

Note that the stronger shift is observed for the LUMO level than the HOMO
level in the acceptor group(s) substituted derivatives, while the trend is reverse in the
donor group(s) substituted derivatives. Similar trends have been explained by a three-
level model including the frontier levels of the PPV repeat unit and the occupied
(unoccupied) molecular orbital of the donor (acceptor) group
13
. We stress that the
extent to which the frontier levels are shifted upon derivatization does depend on the
strength of the coupling between the backbone and the substituents; changing the
nature of the side groups can modulate the band gaps, hence, allows for a fine tuning
of the colour of the emitted light.

5-4 Conclusions

To summarize, we have investigated the influence of electron acceptors and
donors on the geometric and electronic properties of the poly(9,9-dihexylfluorene-1,4-
phenylene) unit cell. We have shown that sizable effects can be obtained. In particular,

95
the use of diamino or dicyano substituents on the phenylene ring allows to decrease
significantly the band gaps. This information is of prime importance in the design of
new chemical structures aimed at a fine tuning of the emitted colour and at a
significant improvement in quantum efficiency.
























96

References

1. Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.;
Friend, R. H.; Burn, P. L. and Holmes, A. B., Nature, 1990, 347, 539.
2. Hide, F.; Díaz-García, M. A.; Schwartz, B. J. and Heeger, A. J., Acc. Chem.
Res., 1997, 30, 430.
3. Greenham, N. C.; Friend, R. H. in Solid State Physics; Enhrenreich, H.;
Spaepen, F. Eds.; Academic Press: San Diego, CA, 1995; Vol. 49, pp 1–149.
4. Gustafsson, G.; Cao, Y.; Treacy, G. M.; Klavetter, F.; Colaneri, N. and Heeger,
A. J., Nature, 1992, 357, 477.
5. Braun, D.; Brown, A. R.; Staring, E. G. J. and Meijer, E. W., Synth. Met., 1994,
65, 85.
6. Greenham, N. C.; Moratti, S. C.; Bradley, D. D. C.; Friend, R. H. and Holmes,
A. B., Nature, 1993, 365, 268.
7. Brown, A. R.; Bradley, D. D. C.; Burroughes, J. H.; Friend, R. H.; Greenham,
N. C.; Burn, P. L.; Holmes, A. B. and Kraft, A., Appl. Phys. Lett., 1992, 61, 23.
8. AMPAC 6.51 Program, Semichem, 7128 Summit, Shawnee, KS 66216.
9. Helbig, M. and Hörhold, H. H., Makromol. Chem., 1993, 194, 1607.
10. Nguyen, T. Q.; Yee, R. Y. and Schwartz, B. J., J. Photochem. Photobiol. A,
2001, 144, 21.
11. Xiao, Y.; Yu, W. L.; Chua, S. J. and Huang, W., Chem. Eur. J., 2000, 6(8),
1318.
12. Halliday, D. A.; Burn, P. L.; Bradley, D. D. C.; Friend, R. H.; Gelsen, O. M.;
Holmes, A. B.; Kraft, A.; Martens, J. H. F. and Pichler, K., Advan. Mater.,
1993, 5, 40.



97
13. Cornil, J.; Beljonne, D.; dos Santos, D. A. and Brédas, J. L., J. Phys. Chem.,

1995, 99, 5604.