Durham Research Online
Deposited in DRO:
13 September 2010
Version of attached le:
Published Version
Peer-review status of attached le:
Peer-reviewed
Citation for published item:
Monkman, A. P. and Burrows, H. D. and Hartwell, L. J. and Horsburgh, L. E. and Hamblett, I. and
Navaratnam, S. (2001) 'Triplet energies of pi-conjugated polymers.', Physical review letters., 86 (7). pp.
1358-1361.
Further information on publisher's website:
http://dx.doi.org/10.1103/PhysRevLett.86.1358
Publisher's copyright statement:
c
2001 by The American Physical Society. All rights reserved.
Additional information:
Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for
personal research or study, educational, or not-for-prot purposes provided that:
•
a full bibliographic reference is made to the original source
•
a link is made to the metadata record in DRO
•
the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full DRO policy for further details.
Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom
Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971
https://dro.dur.ac.uk
VOLUME
86, NUMBER 7 PHYSICAL REVIEW LETTERS 12F
EBRUARY
2001
Triplet Energies of p -Conjugated Polymers
A. P. Monkman,
1
H. D. Burrows,
2
L. J. Hartwell,
1
L. E. Horsburgh,
1
I. Hamblett,
3
and S. Navaratnam
3
1
Department of Physics, University of Durham, Durham, United Kingdom
2
Department of Chemistry, University of Coimbra, 3049 Coimbra, Portugal
3
Paterson Institute for Cancer Research, Christie Hospital, Manchester, United Kingdom
(
Received 4 April 2000)
Using pulse radiolysis and triplet energy transfer has enabled us to measure the triplet energies in a
broad range of different p-conjugated polymers. In all cases we find that the 1
3
B
u
is of order 0.6 to
1 eV below the 1
1
B
u
, indicative of localized triplet states with strong electron-electron correlation. We
also observe that the 1
1
A
g
-1
3
B
u
gap decreases linearly as the 1
1
A
g
-1
1
B
u
gap decreases even though
polymers with very different structure have been studied. This surprising result suggests that polymers
with singlet gap ,1.3 eV will have a triplet ground state.
DOI: 10.1103/PhysRevLett.86.1358 PACS numbers: 78.66.Qn, 71.20.Rv, 71.35.–y
Luminescent conjugated polymers are a new technologi-
cally important class of materials which will be used in
light emitting display devices for the next generation of
information technology based consumer products [1,2].
Their photophysical properties are complex, with many
different excited states and other transient species having
been observed over various time domains [3,4]. However,
the nature and photophysics of triplet states in these materi-
als is not well characterized. This is an important omission
since triplet states are generally believed to be the domi-
nant species formed on charge recombination which yields
electroluminescence, the basis for both polymer light emit-
ting diodes and potential electrically pumped polymeric
lasers. This is because charge recombination spin statistics
dictate that only ca. one singlet exciton is formed for every
three triplet excited states (excitons) [5]. Thus, to be able to
produce better materials, more efficient devices, and gain
a deeper insight into these polymers, the basic characteri-
zation of their triplet states is vital. Further, by comparing
the relative energies of the singlet and triplet “gaps” an
estimate of the strength of electron-electron correlation in
these polymers can be made. In the case of strong correla-
tion, the exciton model would be favored over a band pic-
ture as the description of these materials. However, such
measurements are not trivial, due, in general, to the very
low intersystem crossing rates in these materials [6–8] and
their low phosphorescence quantum yields [9]. Hence via
direct optical excitation it has not generally been possible
to measure triplet energies [10]. Therefore, we have turned
to an alternative route to produce triplet states in these
polymers, one in which both the kinetics of the triplets
and more importantly their energies can be measured.
We recently reported on the use of pulse radiolysis to
measure the triplet energy in poly[2-methoxy,5-(2
0
-(ethyl-
hexoxy)-p-phenylenevinylene] (MEHPPV) [7]. We have
also demonstrated that charged states, again generated by
this method, have very different lifetimes and spectral sig-
natures from the triplet states [11]. We now report on
triplet state measurements on a range of the most widely
used luminescent and conductive polymers in current dis-
play technologies; see Fig. 1. In general we find, as in
acenes, the energy of the first triplet state, 1
3
B
u
,tobe
typically
2
3
that of the first singlet 1
1
B
u
excited state, in-
dicative of strong electron-electron correlation.
Since the classic work of Terenin and Ermolaev [12],
triplet-triplet energy transfer has become one of the most
important methods of specifically generating triplet states
of organic molecules. Coupling this with pulse radioly-
sis of solutes in organic solvents provides an excellent
technique for the selective creation and study of both
excited states and charged species [13–15]. An intense,
10–100 ns pulse of electrons (accelerated to 10 MeV)
from a linear accelerator irradiates the sample. The radia-
tion chemistry of solutions is dominated by the most
prevalent species in solution, the solvent [13]. The
primary excitation process involves ionization which
in aromatic solvents such as benzene (BZ) is rapidly
followed by charge recombination 共,1 ns兲, leading to
excited triplet and singlet state generation in the statistical
ratio 3:1 (in the benzene). By using appropriate energy
acceptors 共A兲, such as biphenyl, having lower triplet
energies than benzene, short-lived singlet excited states
and high S
1
! T
1
intersystem crossing efficiencies, the
excited triplet state of this acceptor can be selectively
produced and subsequently transfer the triplet energy to
the molecule under investigation 共S兲, i.e., our polymers,
subject to the kinetically demanded concentration ratio
关Bz兴 ¿ 关A兴 ¿ 关S兴 [16]. This concentration gradient,
combined with diffusion controlled collisional triplet
energy transfer (in solution) ensures that effectively only
triplet energy is transferred down to the molecule under
investigation. Transient induced optical absorption spectra
are used to monitor triplet state grow-in and decay and
ground state bleaching. We have on the order of 20
appropriate acceptors/donors available to us [17] with
well characterized T
1
energies E
TA
, each within 0.1 to
0.2 eV of the next. A series of measurements are made
to determine which acceptor transfers triplet energy to the
polymer. When E
TA
, E
TS
no induced triplet absorption
is observed since no triplet transfer to our polymer is
1358 0031-9007兾01兾86(7)兾1358(4)$15.00 © 2001 The American Physical Society
VOLUME
86, NUMBER 7 PHYSICAL REVIEW LETTERS 12 F
EBRUARY
2001
S
S
S
S
n
S
S
S
S
n
N
N
N
N
n
N
N
N
N
n
O
O
O
O
O
O
O
O
n
O
O
O
O
O
O
O
n
O
O
O
O
O
O
O
O
O
n
N
N
N
N
O
O
O
O
O
O
O
n
O
O
O
OO
S
SS
S
n
N
N
N
N
n
H
H
n
poly(2,5-pyridinediyl)
(PPY)
poly(3-hexyl-2,5-pyridinediyl)
(HPPY)
poly(3-octylthiophene)
(P3OT)
poly(3-octyl-4-methylthiophene)
(PMOT)
poly(2,5-hexyloxyphenylenevinylene)
(DHOPPV)
poly(2-methoxy-5-(2'-ethylhexyoxy)-p-phenylenevinylene)
(MEHPPV)
poly(2-butyloxy-5-octylphenyl
-3-thiophene) (PBOPT)
poly(dioctylflourene)
(PFO)
poly(2,5-octyloxyphenylenevinylene)
(DOOPPV)
poly(2-methoxy-5-(2'-ethylhexyoxy)
-p-phenylenecyanovinylene)
(CN-MEHPPV)
polyemeraldine
(PANi)
FIG. 1. Schematic pictures of the chain structures of the polymers used in this work and the acronyms used in Table I and the text.
possible, thus enabling an upper limit to be placed on the
triplet energy of the polymer. For the range of acceptors
available to us we can estimate the triplet energy of the
polymer to within 60.05 to 0.1 eV in most cases.
To verify the validity of this technique, a comparison
to a direct optical measurement of a conjugated polymer
triplet state is needed. Recently Romanovskii et al. were
able to observe weak phosphorescence from the ladder
polyparaphenylene, MeLPPP [9]. At 77 K in the solid
state they determine a triplet energy of 16 500 cm
21
,
2.05 eV. We have now made a radiolysis experiment,
in benzene solution at room temperature, on MeLPPP
yielding a value of 2.15 6 0.1 eV. Given the small
difference between absorption and emission in solution
compared to solid state in MeLPPP this agreement is very
satisfactory and now gives strong support for the radiolysis
methodology. More details on these measurements are
reported in a separate publication [18].
TABLE I. Photophysical data obtained by pulse radiolysis for triplet states in various conjugated polymers; DE
ST
is the energy
difference between the 1
1
Bu and the 1
3
Bu states.
S
0
-S
1
energy (eV) T
1
-T
n
DE
ST
共l兾nm兲
T
1
-S
0
(eV)
t
T
(eV)
Polymer Maximum
b
Onset
c
PL
(eV)
共l兾nm兲
(ms
a
) Maximum
b
Onset
c
PMOT 3.77 3.22 2.64 2.20 1.85 62 1.57 1.02
(329) (385) (469) (670)
P3OT 2.83 2.42 2.21 1.65 1.50 21 1.03 0.62
(438) (512) (562) (825)
PBOPT 2.52 2.18 2.17 1.60 1.38 57 0.92 0.58
(492) (570) (572) (900)
MEHPPV 2.48 2.23 2.25 1.30 1.50 92 1.18 0.93
(500) (556) (550) (830)
PFO 3.22 3.01 3.00 2.30 1.65 108 0.92 0.71
(385) (412) (413) (750)
DOOPPV 2.59 2.23 2.28 1.50 1.55 134 1.09 0.73
(478) (556) (544) (800)
DHOPPV 2.58 2.13 2.30 1.50 1.62 176 1.08 0.63
(480) (581) (538) (765)
PPY 3.35 3.10 3.26 2.40 2.10 98 0.95 0.70
(370) (400) (380) (590)
HPPY 3.90 3.50 2.82 2.50 2.10 70 1.40 1.00
(318) (344) (440) (590)
CN-MEHPPV 2.72 2.35 2.28
(456) (527) (544)
PANi
d
2.00 1.77 N/A ,0.9 1.55 ca. 2.4 $ 2000 ,1.1 ,0.9
(700) (800)
a
Triplet state lifetime in benzene solution;
b
from maximum of absorption band;
c
from onset of absorption band;
d
measured in 80%
benzene/20% N-methylpyrrolidone.
1359
VOLUME
86, NUMBER 7 PHYSICAL REVIEW LETTERS 12 F
EBRUARY
2001
The results of our triplet energy measurements are sum-
marized in Table I. Triplet lifetimes in benzene solution
are also given. It can be noted that the lifetimes for the
polythiophenes are markedly shorter than those of the
other polymers. This is in agreement with observations
on triplet states of oligothiophenes [19] and time resolved
induced absorption measurements on P3OT [20] and is
consistent with significant spin orbit coupling involving the
sulfur atom in each repeat unit increasing intersystem
crossing. For CN-MEHPPV we find that a metastable
transient species (either an isomer or degradation product)
appeared to have formed via the triplet state. Only an
upper limit of ,2.6 eV can be made for the triplet energy
in this case. All of the polymer triplet states except that
of polyaniline emeraldine base (PANi) were quenched by
oxygen. Thus, for PANi it was possible only to estimate
an upper limit on the T
1
energy.
To determine if there are any general trends between
singlet and triplet energies and polymer structure, in
Fig. 2 we plot the 1
1
A
g
-1
3
B
u
共S
0
-T
1
兲 energy gap against
the 1
1
A
g
-1
1
B
u
共S
0
-S
1
兲 separation, as measured from
the maximum of the singlet absorption band. A very
striking linear correlation between the singlet and triplet
energies is found for the very different polymers studied
here. This is a very unexpected result. We note that a
similar trend can be found with the acenes [21]. As a
first approximation, for the conjugated polymers a least
squares fit to the data yields
T
1
苷 共1.13S
1
2 1.43兲 6 0.25 eV .
Several important points can be deduced from this find-
ing. The triplet gap decreases along with the reduction
in singlet gap. By extrapolation it is possible to predict
that a polymer will have a triplet ground state when the
singlet energy gap is ca. 1.3 6 0.25 eV. For comparison,
with polyacenes it is suggested that the first member with
a triplet ground state will be nonacene [21]. From litera-
ture data, we estimate an energy for its lowest singlet state
to be ca. 1.3–1.5 eV. The corresponding energies for the
heterocyclic polyacenes may be even lower [22]. Thus the
conjugated polymers would seem to behave in an identical
fashion to the polyacences and by comparison the conju-
gation length in polymers must be ca. nine repeat units.
The reduction of the singlet gap, i.e., the decrease in sin-
glet exciton energy, as conjugation length increases is gen-
erally accepted to imply that increased conjugation length
delocalizes excitations on the polymer chain. This is borne
out in many experiments on oligomeric compounds; for
example, see Refs. [23] and [24], where it is found that
the DE
S
共L兲 苷 DE
S
共`兲 1 C
S
兾L, where DE
S
is the singlet
gap, L is the chain length, and C
S
is a constant [23]. As the
triplet states are essentially made up of the same orbitals
as the singlets, a similar relationship should hold for the
triplet states; i.e., DE
T
共L兲 苷 DE
T
共`兲 1 C
T
兾L. For the
case where C
T
, C
S
the singlet-triplet gap will decrease
as L increases. In our experiments, each backbone is dif-
ferent; thus the values of the constants in these relation-
FIG. 2. Plot of triplet energy gap against singlet energy gap
for different polymers. Polymer acronyms given in Fig. 1. The
line is a guide to the eye.
ships will be different. However, the general trend that we
observe for this broad range of conjugated polymers is that
the triplet gap decreases as the singlet gap decreases, i.e.,
delocalization increases, implying that C
T
, C
S
is true for
conjugated polymers in general. Although we have studied
only three polythiophene derivatives, they do fit on a good
straight line; in these cases the backbone is the same and
only the conjugation length is different, caused by torsional
distortion from the sidegroups [25]. We are in the near
future going to measure more of this family of polymers
to verify these initial trends. The physical consequence
of C
T
, C
S
is that the triplet states must be more tightly
bound than the singlet states, i.e., more localized, but their
degree of localization is affected by the chain backbone.
This general conclusion is in agreement with the observa-
tion of vibronic structure on the triplet-triplet absorption
band, indicating a structurally relaxed triplet state [7,17].
Theoretical predictions [26–28], work on model systems
[9,10], and the observation of long-lived phosphorescence
in a ladder poly(phenylene) [29] are also consistent with
localized triplet states. In addition, it is difficult to ex-
plain the relatively low triplet mobility on these polymer
chains, or the observation of delayed fluorescence follow-
ing triplet-triplet annihilation [18], unless they are more
tightly bound than the corresponding singlet states.
The large electron-electron correlation seen in all these
polymers implies that the on site repulsion energy U is
large. Abe et al. have shown that for the 1
3
B
u
to be ap-
preciably lower than the 1
1
B
u
, the ratio U兾V . 2, where
V is the long range Coulomb interaction [30]. In the case
where U is large, the triplet electrons can be in the same
molecular orbital with a minimum of repulsion. This again
is consistent with localized triplet states. In the solid state
this scenario would strongly support the exciton picture
over the band model. As we find the same triplet energy
1360
VOLUME
86, NUMBER 7 PHYSICAL REVIEW LETTERS 12 F
EBRUARY
2001
in solution as is found in solid state for MeLPPP and also
find very little difference between triplet-triplet absorption
energies in solution and solid state, we assume that there
is little change in the strong electron-electron correlation
when weak interchain interactions are introduced in the
solid state.
To conclude, we have now measured the triplet ener-
gies of a wide range of conjugated polymers, in the main
those predominant in the areas of light emitting devices.
Comparing these energies with the corresponding singlet
energy it can be seen that in all cases the triplet state is at
a much lower energy than the singlet indicative of a large
electron-electron correlation energy. A general trend is ob-
served between the singlet and triplet energy gaps. For all
the various types of conjugated polymers studied here there
exists a linear relationship between singlet and triplet gaps.
The confirmation that triplet energies determined by our
method yield very similar values to those found by phos-
phorescence in the solid state allows the triplet energies
given here to be used as a good guide to the triplet ener-
gies in the solid state as well as the isolated chain case. Full
understanding of electron correlation in the excited states
of these conjugated polymers requires detailed quantum
mechanical treatment, but the experimental data presented
on triplet energies represent both a challenge to theoreti-
cians and a good starting point for high level calculations.
One very obvious initial prediction from this work is that
a polymer with a singlet gap below ca. 1.3 eV should have
a triplet ground state.
We acknowledge the EPSRC (GR/M86040), PRAXIS
XXI (Project No. 2/2.1/QUI/411/94), and the British
Council/CRUP (Project No. B-9/97) for financial sup-
port. We thank and acknowledge for their very kind
supply of materials Covion (MEH-PPV), Professor M.
Andersson (polythiophenes), Professor U. Scherf (PFO),
M. de Long (DOO and DHO PPV), and Professor A.
Holmes (CN-MEHPPV). We also thank the Paterson
Institute for Cancer Research Free Radical Research Fa-
cility (Manchester, UK) for access to the pulse radiolysis
spectrometer, Dr. Donald Allan for all of his help and
Scottish history lessons, and Professor H. Bässler for
enlightening discussions.
[1] R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes,
R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. dos Santos,
J. L. Brédas, M. Lögdlund, and W. R. Salaneck, Nature
(London) 397
, 121 (1999).
[2] R. J. Visser, Philips J. Res. 51
, 467 (1998).
[3] N. T. Harrison, G. R. Hayes, R. T. Phillips, and R. H.
Friend, Phys. Rev. Lett. 77
, 1881 (1996).
[4] S. V. Frolov, W. Gellermann, M. Ozaki, W. Gellermann,
K. Yoshino, and Z. V. Vardeny, Phys. Rev. Lett. 78
, 729
(1997).
[5] M. A. Baldo, D. F. O’Brien, M. E. Thompson, and S. R.
Forrest, Phys. Rev. B 60
, 14 422 (1999).
[6] R. D. Scurlock, B. Wang, P. R. Ogilby, J. R. Sheats, and
R. L. Clough, J. Am. Chem. Soc. 117
, 10 194 (1995).
[7] A. P. Monkman, H. D. Burrows, M. da G. Miguel, I. Ham-
blett, and S. Navaratnam, Chem. Phys. Lett. 307
, 303
(1999).
[8] J. Seixas de Melo, C. S. Soare, A. P. Monkman, H. Bur-
rows, and L. G. Arnaut (to be published).
[9] Yu. V Romanovskii, A. Gerhard, B. Schweitzer, U. Scherf,
R. I. Personov, and H. Bässler, Phys. Rev. Lett. 84
, 1027
(2000).
[10] K. A. Walters, K. D. Ley, and K. S. Shanze, Chem. Com-
mun. No. 10, 1115 (1998).
[11] H. D. Burrows, M. da G. Miguel, A. P. Monkman, L. E.
Horsburgh, I. Hamblett, and S. Navaratnam, J. Chem. Phys.
112
,
3082 (2000).
[12] A. N. Terenin and V. Ermolaev, Trans. Faraday Soc. 52
,
1042 (1956), and references therein.
[13] J. H. Baxendale and M. A. J. Rodgers, Chem. Soc. Rev. 7
,
235 (1978).
[14] A. J. Swallow, in Radiation Chemistry, Principles and
Applications, edited by Farhataziz and M. A. J. Rodgers
(VCH, New York, 1987).
[15] J. H. Baxendale and M. Fiti, J. Chem. Soc. Faraday Trans.
2 68
, 218 (1972).
[16] E. J. Land, Proc. R. Soc. London A 305
, 457 (1968);
R. Bensasson and E. J. Land, Trans. Faraday Soc. 67
, 1904
(1971).
[17] A. P. Monkman, H. D. Burrows, L. E. Horsburgh, L. J.
Hartwell, M. da G. Miguel, I. Hamblett, and S. Navarat-
nam, Proc. SPIE Int. Soc. Opt. Eng. 3797
, 109 (1999).
[18] A. P. Monkman, H. D. Burrows, I. Hamblett, S. Navarat-
nam, U. Scherf, and C. Schmitt, Chem. Phys. Lett. 327
,
111–116 (2000).
[19] J. Seixas de Melo, L. M. Silva, L. G. Arnaut, and R. S.
Becker, J. Chem. Phys. 111
, 5427 (1999).
[20] B. Kraabel, D. Moses, and A. J. Heeger, J. Chem. Phys.
103
, 5102 (1995).
[21] H. Angliker, E. Rommel, and J. Wirz, Chem. Phys. Lett.
87
, 208 (1982).
[22] Th. Starssner, A. Weitz, J. Rose, F. Wudl, and K. N. Houk,
Chem. Phys. Lett. 321
, 459 (2000).
[23] J. Grimme, M. Kreyenschmidt, K. Müllen, and U. Scherf,
Adv. Mater. 7
, 292 (1995).
[24] H. S. Woo, O. Lhost, S. C. Grahem, D. D. C. Bradley,
R. H. Friend, C. Quattrocchi, J-L. Brédas, R. Schenk, and
K. Müllen, Synth. Met. 59
, 13 (1993).
[25] M. R. Andersson, O. Thomas, W. Mammo, M. Svensson,
M. Theander, and O. Inganäs, J. Mater. Chem. 9
, 1933
(1999).
[26] D. Beljonne, Z. Shaui, R. H. Friend, and J-L. Brédas,
J. Chem. Phys. 102
, 2042 (1995).
[27] M. Chandross and S. Mazumdar, Phys. Rev. B 55
, 1497
(1997).
[28] D. A. dos Santos, D. Beljonne, J. Cornil, and J. L. Brédas,
Chem. Phys. 227
, 1 (1998).
[29] Yu. V. Romanovskii and H. Bässler, Chem. Phys. Lett.
326
, 51 (2000).
[30] S. Abe, J. Lu, and W. P. Su, Phys. Rev. B 45
, 8264 (1992).
1361
