Exciton Dynamics in Hybrid Polymer/QD Blends

Abstract

The prospect of exploiting quantum dots (QDs) properties (tunable absorption spectrum, multiple exciton generation) while maintaining the flexible structure of polymer systems opens new possibilities in the photovoltaic field. Although charge transport dynamics in pristine polymer and QDs systems have been quite well established lately, a complete understanding of the charge transfer process between QDs and polymers when they are in blends is still lacking. In this work we used static and ultrafast fluorescence spectroscopy together with Atomic force Microscopy (AFM) to study the exciton dynamics in polymer/QDs films. Specifically we used poly(3-hexylthiophene) (P3HT) as the hole conducting donor material and the core shell CdSe(ZnS) QDs as the electron acceptor material. The QDs surface has been treated with two different capping ligands treatments: one based on the use of pyridine and the other one on hexanoic acid. The influence of the two different methods on the exciton dynamics and on the morphology will also be discussed. Blends containing differently treated P3HT/CdSe(ZnS) wt% ratios have been prepared producing films having uniform morphology and good intermixing, as proved by AFM measurements. Ultrafast fluorescence decays allowed us to compare the exciton dynamics in the polymer pristine respect to the treated P3HT/CdSe(ZnS) films. Efficient fluorescence quenching has been shown by both kind of blends respect to the pure polymer.

Full text

Energy Procedia 44 ( 2014 ) 167 – 175
1876-6102 © 2013 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of The European Materials Research Society (E-MRS)
doi: 10.1016/j.egypro.2013.12.024
ScienceDirect
E-MRS Spring Meeting 2013 Symposium D - Advanced Inorganic Materials and Structures for
Photovoltaics, 27-31 May 2013, Strasbourg, France
Exciton Dynamics in Hybrid Polymer/QD Blends
Annalisa Brunoa
*
, Tiziana Di Luccioa, Carmela Borrielloa, Fulvia Villania, Saif
A. Haqueb, Carla Minarini a
aItalian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), p.le E. Fermi 1, 80055 Portici (NA),
Italy; bChemestry Department, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
Abstract
The prospect of exploiting quantum dots (QDs) properties (tunable absorption spectrum, multiple exciton generation) while
maintaining the flexible structure of polymer systems opens new possibilities in the photovoltaic field. Although charge
transport dynamics in pristine polymer and QDs systems have been quite well established lately, a complete understanding of
the charge transfer process between QDs and polymers when they are in blends is still lacking. In this work we used static and
ultrafast fluorescence spectroscopy together with Atomic force Microscopy (AFM) to study the exciton dynamics in
polymer/QDs films. Specifically we used poly(3-hexylthiophene) (P3HT) as the hole conducting donor material and the core
shell CdSe(ZnS) QDs as the electron acceptor material. The QDs surface has been treated with two different capping ligands
treatments: one based on the use of pyridine and the other one on hexanoic acid. The influence of the two different methods on
the exciton dynamics and on the morphology will also be discussed. Blends containing differently treated P3HT/CdSe(ZnS)
wt% ratios have been prepared producing films having uniform morphology and good intermixing, as proved by AFM
measurements. Ultrafast fluorescence decays allowed us to compare the exciton dynamics in the polymer pristine respect to the
treated P3HT/CdSe(ZnS) films. Efficient fluorescence quenching has been shown by both kind of blends respect to the pure
polymer.
© 2013 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of The European Materials Research Society (E-MRS).
* Corresponding author. Tel.:+390817723378.
E-mail address: annalisa.bruno@enea.it; tiziana.diluccio@enea.it
Available online at www.sciencedirect.com
© 2013 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of The European Materials Research Society (E-MRS)

168 Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175
Keywords: Quantum Dots; hybrids; ligand exchange; ultrafast fluorescence
1. Introduction
Organic and hybrid QDs based solar cells belong to the 3rd generation photovoltaics that try to overcome the limits
of traditional mainly based on expensive silicon technology. Organic based solar cells are relatively cheap to
fabricate using inexpensive, simple coating processes such as inkjet printing or spin coating. Besides Cheap
fabrication, organic and hybrid photovoltaics offers other attractive features, such as flexibility, processability and
tunability of the optical band gap [1].
Hybrid cells can be realized by incorporating the QDs in polymers to form bulk heterojunction cells, where the
polymer is usually the electron donor and the dots are the electron acceptor material.
Hybrid blends for organic electronics combine advantages of both organics, as easy processability, flexibility,
cheap material, and inorganic QDs, good transport proprieties, tuneable band gap, multiple exciton generation
(MEG). In particular many are the advantages of employing semiconducting QDs, due to quantum confinement
effects that can be easily controlled by consolidated synthesis routes. One of the most relevant phenomena that
different types of quantum dots have shown is MEG that could greatly enhance the power conversion efficiency up
to 66% [2].
A great advantage of hybrid systems is that QD band gap can be properly tuned with the dot size and adapted to the
specific polymer used in order to maximize the light absorption [3-5]. Moreover, in bulk heterojunction carrier
recombination is strongly reduced respect to p-n junction thanks to the high number of acceptor-donor interfaces.
Both aspects, wider absorption spectrum and reduced losses of carriers, contribute to improve the overall cell
efficiency.
As mentioned before hybrid cells allow employing large area solution processing and low cost materials. All these
advantages have potentially a very relevant societal impact addressing the important issue of energy harvesting.
Nevertheless a number of major challenges have to be solved to obtain a successful hybrids solar cell based on
polymer/quantum dots composite system. A suitable choice of the ligand surrounding the QDs is critical to
disperse them homogeneously in the polymer matrix but also to maximize the charge separation and transport in
the composite. Indeed from synthesis QDs are stabilized in solution by long alkyl chains that are responsible for
low conductivity when the QDs/polymer blends are deposited in films.
Surface treatments of the QDs are needed to replace the original long alkyl chain ligands with shorter molecules
that improve the carrier mobility among dots and ultimately the device efficiency.
The systems we are currently studying is based on P3HT polymer and luminescent CdSe(ZnS) QDs. We have
replaced the original ligand, octadecylamine (CH3(CH2)17NH2), through an exchange process with pyridine and
with an hexanoic acid (HA) both devoted to shorten the insulating shell around CdSe(ZnS) QDs [6-8].
The aim of this work is to investigate the effect of the two different ligands exchange processes on the excitons
dynamics in the P3HT/CdSe(ZnS) blends studying the exciton fluorescence decays for different concentrations of
the treated QDs inside the blends. Moreover the morphology and QDs cluster formation in the films have been
analysed by Atomic Force Microscopy (AFM).
2. Experimental
2.1. Materials
Pyridine, hexane and methanol were purchased by Carlo Erba reagent, hexanoic acid, chlorobenzene, and P3HT
by Sigma Aldrich. All reagents mentioned have been used as received.
The P3HT energy levels of valence (HOMO) and conduction (LUMO) have been previously measured to be
3.2 eV and -5.2 eV [ 9] respectively.CdSe(ZnS) HOMO level is -3.2 eV and -7.4 eV [10].
The core shell CdSe(ZnS) QDs, absorbing at 640 nm, surrounded by octadecylamine have been purchased by
N,N-Labs and treated to remove the native ligand using two different ligand treatments.

Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175 169
In the first process we exchanged octadecylamine by pyridine. In this procedure, 40 mg of QDs were
suspended in pyridine at the concentration of 10 mg/mL and sonicated at 50 °C for 3 hours. After about 10 minutes
QDs were dissolved in pyridine. QDs were precipitated and washed by adding hexane. The whole procedure was
repeated two more times to obtain pyridine capped CdSe(ZnS). QDs were dissolved at the concentration of 17
mg/mL in a mixture chloroform/pyridine containing about 16% in volume of pyridine. The QDs were
characterized by UV-visible and photoluminescence spectroscopy.
In the second process 40 mg of QDs were dissolved in hexanoic acid (0.5 mg/mL) and heated at 105° C for 10
minutes. While the solution was cooling down, QDs precipitation started but it was completed by adding methanol.
The residue was recovered by centrifuge (8000 rpm for 20 minutes) and washed by methanol. Finally hexanoic
acid treated QDs were dissolved in chloroform at the concentration of 20 mg/ml. The optical properties of QDs
before and after both surface treatments were characterized by UV-visible and photoluminescence spectroscopy.
In this method the acid doesn’t remain in the coordination sphere of QDs but it forms the ammonium salt of
octadecylamine. This salt is easily removed by QDs purification reducing insulating sphere around QDs[6].
Blends P3HT/QDs were prepared by mixing in chloroform P3HT and treated QDs in the appropriate ratios to
obtain blends containing 10, 25, 50, 75 % wt of QDs for both treatments methods. The blends were spin coated on
glass substrates at 1000 rpm for 30 s. The thickness of the samples was measured using an optical profilometer.
The pristine sample was 120 nm tick and the tickeness of the different concentrations wwas varying from 100 to 70
nm.
All the blend films have been characterized by UV-visible absorption, static and time resolved
photoluminescence emission spectroscopy and Atomic Force microscopy (AFM).
2.2. Methods
The blends films were characterized by UV-Vis absorption spectroscopy by a Perkin Elmer Lambda 900
Spectrophotometer.
Photoluminescence emission of the pure polymer and the blends deposited by spin coating on quartz substrate has
measured by Fluorolog 3 instrument by Horiba Jobin Instruments SA.
The ultra-fast fluorescence emission experiments were performed using a femtosecond laser based systems. The
Second Harmonic output of a mode-locked Ti:Sapphire oscillator with the wavelength fixed at 800 nm as
excitation beam was used as excitation beam. The pulse duration was 70 fs and the repetition rate 80 MHz. A
portion of this fundamental beam was frequency doubled to create the excitation beam with a wavelength of 400
nm.
Fluorescence from the sample was focused on a beta barium borate (BBO) crystal along with the 800 nm
fundamental beam. The fluorescence emitted at different emission wavelength was mixed in the BBO crystal with
the gate beam to generate sum frequency photons which were detected using a photomultiplier tube. Data were
acquired using Lab-View software and subsequently analyzed.
The temporal reconstruction of the signal is obtained through a micrometrical precise delay line on the gate beam.
The resolution of the system was measured to be 150 fs.
Sample degradation was avoided by performing the measurements under flowing nitrogen and using a translation
stage to move the sample within the beam, removing the effect of photo bleaching and providing data averaged
across the whole of the sample. The system has been also previously presented in previous works [11].
The surface morphology of the polymer-fullerene blend layers was also analyzed by atomic force microscopy
(AFM, Veeco, Dimension Digital Instruments Nanoscope IV) in tapping mode configuration.
3. Results
In order to realize a highly-interconnected pathway in the P3HT/QDs blends the slightly insulating, long aliphatic
ligands used to passivate the QD surfaces have to be removed.
3.1 Pyridine exchange

170 Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175
Herein the first QDs investigated were ligand exchanged with pyridine. This method has already shown the
reduction of inter-particle distances which enhanc the charge transport [6]. Absorption spectra of pristine P3HT
and P3HT/QDs hybrid films were measured in the UV-visible range for all the QDs range of concentration
investigated. The steady state absorption spectra for all the pyridine capped CdSe(ZnS) QDs concentration in
P3HT/QDs hybrid films are displayed in Figure 1 a). The Absorption shape remains unchanged increasing QDs
concentration although a total reduction of the absorption spectra can be observed as expected due to the smaller
P3HT amount in the blends. Figure 1 b) shows the emission spectra after excitation at 400 nm for the same films
with different concentrations of QDs in the blends. It is clear that the fluorescence is quenched more efficiently the
more QDs are added to the blends. The broad peak due to the contribution of QDs emission is clearly visible
around 650-660 nm, invresong the QDs concentration in the blend.
Ultrafast Fluorescence decays for the Pure P3HT and all P3HT/QDs blends, using an excitation beam at 400 nm
and collected at 650 nm are reported in Figure 2 a). The fluorescence lifetime give a measure the exciton diffusion
lifetime in the pure polymer and in the blends.
We can observe that pristine fluorescence decays is well described by a double exponential decays as expected
for regio regular P3HT films [11-13]. A fast (<3 ps) and slow (>3 ps) time decays are present for all the blends
compositions. The fast-phase dynamics remain constant over all blend compositions, whereas the slow-phase time
decay decreases upon increasing the relative content of quencher. The fast decay is related to nonradiative
processed like tortional arrangement of the exciton inside the polymer chain [10].
Figure 1: Steady-state absorption spectra a) and Emission spectra of hybrid films of P3HT/ with different weight
ratios of CdSe(ZnS): (black) 0 wt%, (red) 10 wt%,(blue) 25 wt%, (dark green ) 50 wt% and (pink)75 wt%.
Figure 2: P3HT: QDs (pyridine exchanges) films: Fluorescence decays collected at 650 nm after excitation at 400 nm a) and extrapolated
lifetimes as function of QDs concentration b).
300 400 500 600 700
0,0
0,4
0,8
1,2
P3HT
10%
25%
50%
75%
Absorption
Wavelength/ nm
550 600 650 700 750
0,0
5,0x10
6
1,0x10
7
1,5x10
7
2,0x10
7
2,5x10
7
3,0x10
7
P3HT
10%
25%
50%
75%
Wavelength/ nm
Emission Corrected
-10 0 10 20 30 40 50 60 70 80
0
20
40
60
80
100
120
140
Time (ps)
NC %
050 100
0.0
0.2
0.4
0.6
0.8
1.0
Time/ps
Normalized fluorescence
P3HT
10 %
25 %
50 %
75 %

Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175 171
The fact that the fast-phase decay remains constant over all blend compositions suggests that the nonradiative
decay pathways are similar for all acceptors composition. The slow time decay is instead related with the diffusion
of the excite state and so it is correlated with the presence of different acceptors concentrations. The fluorescence
kinetics of the blends show a similar time trend respect to the pristine films but at the same time an evident
reduction of lifetimes increasing the concentration of QDs in the blends. The life time values as function of the dots
concentration is reported in Figure 1 b). The strong quenching effect is evident already with concentrations as small
as 10 wt% in the blends.
In Figure 4 a), 2-dimensional(2D)-height and phase AFM images of the surfaces of P3HT/CdSe(ZnS) blend sample
are reported. Along the scanned area (scan size 1 x 1 mm2), the morphological analysis indicated good intermixing
of polymer and QDs in the spin-coated films characterized by uniform surfaces with root-mean-square roughness
(Rrms) values around 8 nm, that is very good for hybrids blends, still a bit higher compared to organic
P3HT/PCBM mixtures used in solar cells [14].
A small value of the roughness is an index that the good phase separation between P3HT and QDs occurred
[15]. Nevertheless, the morphological analysis of P3HT/CdSe(ZnS) films shows clearly the presence of
nanoclusters that have been observed all over the sample. The average size of these nanoclusters has been
estimated, through section analysis, to be around 30 nm, as shown in Figure 3 b).
Efficient charge photogeneration requires the morphology of the donor–acceptor film to be structured on the
exciton diffusion length of the donor material (typically on the order of 15 nm). This condition is necessary to
ensure that photogenerated excitons are in proximity (within the exciton diffusion length) of the donor acceptor
heterojunction where they can undergo dissociation into electrons and holes. Moreover we would like to point that
both the roughness and the esteem of the domain sizes has been done in different areas of the sample. The
roughness and the sizes of the domains resulted comparable.
Figure 3: P3HT: QDs ( pyridine exchanged) films: 2D-height and phase AFM images a) and Section Analysis b) ;

172 Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175
In Figure 4 we report a AFM image of the pure P3HT were it where we can observe that the roughness of the film
is much smaller (Rrms) values around 1.6 nm and the surface more homogeneous (observable from the phase
image ) respect to the blended films.
3.1 Hexanoic Acid treatment
CdSe(ZnS) QDs were also treated by hexanoic acid, a method reported recently by Kruger group that have shown
to give very good results in terms of charge transfer in P3HT/CdSe blends [6] indicating the realization of a highly-
interconnected pathway.
The steady state absorption spectra for all the in P3HT/QDs HA treated hybrid films are shown in Figure 5 a). Also
in this case the absorption shape remains unchanged, increasing QDs concentration, although a total reduction of
the absorption spectra can be observed as expected due to the smaller P3HT amount in the films. The characteristic
crystalline shoulders at 550 and 610 nm in this case are less pronounced respect to the films prepared with QDs
exchanged in pyridine. In Figure 5 b) the emission spectra after excitation at 400 nm for films with different
concentrations of QDs in the blends are reported. The quenching of the fluorescence increasing the concentration
of QDs seems to be even more efficient for the QDs treated in HA respect to the ones treated in pyridine.
Figure 5: Steady-state absorption spectra a) and Emission spectra b) of hybrid films of P3HT: with different weight ratios of CdSe(ZnS) HA
treated : (black) 0 wt%, (red) 10 wt%,(blue) 25 wt%, (dark green ) 50 wt% and (pink)75 wt%.
Figure 4: pure P3HT film: 2D-height and phase AFM
300 400 500 600 700
0,0
0,4
0,8
1,2
P3HT
10%
25%
50%
75%
Absorption
Wavelength/ nm
550 600 650 700 750
0,0
5,0x10
6
1,0x10
7
1,5x10
7
2,0x10
7
2,5x10
7
3,0x10
7
P3HT
10%
25%
50%
75%
Wavelength/ nm
Emission Corrected

Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175 173
Ultrafast Fluorescence decays for the Pure P3HT and all P3HT/QDs blends, using excitation at 400 nm and
collected at 650 nm are reported in Figure 6 a). Also in this case P3HT fluorescence decay is well described by a
double exponential decay [11-13]. The fluorescence kinetics of the blends present a double exponential decay and
an evident reduction of the slow lifetimes increasing the concentration of QDs in the blends, as observed for the
QDs treated with pyridine. At the same time the fast component remain constant
Lifetimes as function of the QDs concentration are reported in Figure 6 b). A strong reduction of fluorescence
lifetimes is evident for QDs concentrations grater that 10 wt% in the blends. In this case, respect to the blends
where QDs were exchanged in pyridine, the fluorescence decays are more efficiently quenched for similar QDs
concentration value. This indicate that the excitons can freely move inside the polymer for a shorter time before
reaching an acceptor, and so more efficiently leads to charge formation. This suggest that this ligand exchange
method could give a better percolation and mixing between the QDs and the polymer and this produce a more
efficient exciton quenching and a charge separation process. Blends morphology has been analyzed using AFM
also in this case: 2 dimensional(2D)-height and phase AFM images of the surfaces of P3HT/QDs HA treated are
shown in Figure 7 a). Also in this case the scan size was 1 x 1 mm2.
The images showed good intermixing of P3HT and QDs in films characterized by a flat Rrms value smaller than 4
nm. Also in this case the roughness and the esteem of the domain sizes has been done in different areas of the
sample.
Nevertheless, the morphological analysis of P3HT/CdSe(ZnS) film showed also in this case clearly the presence
of nanoclusters. The nanoclusters size are about 20 nm indicating a better intermixing with the QDs. So the less
invasive HA treatment seems to be more efficient in induce blended films with a fine mixing morphology.
Figure 6: P3HT:QDs( HA exchanged ) films: Fluorescence decays collected at 650 nm after excitation at 400 nm a) and extrapolated lifetimes
as function of QDs concentration b).
010 20 30 40 50 60 70 80
0
20
40
60
80
100
120
140
Time (ps)
NC %
0510 15 20 25 30 35 40 45 50
0.0
0.4
0.8
P3HT
10
25
50
75
Emission
Time/ps
Normalized fluorescence

174 Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175
4. Conclusions
In this work we have investigated the effect of different ligands exchange processes on CdSe(ZnS) to remove the
long chains on the QDs, on the polymer/QDs blends film morphology and their spectroscopic characteristics. In
particular, pyridine and hexanoic acid (HA) treatments were used and after both ligand exchange treatments, the
QDs have been blended with the P3HT polymer.
Both spectroscopic and the structural investigation has been performed on the P3HT/CdSe(Zns) films, showing a
strong fluorescence effect with both ligands exchange methods used.
Moreover the excitons dynamics in such blends has been studied thought ultrafast fluorescence measurements.
Fluorescence decays have been explored for different concentrations of QDs inside the blends, showing that a
exciton lifetime has a more evident reduction in the blend with higher QDs contents for the blends where the QDs
were HA treated.
Both morphologic and spectroscopic methods showed that HA treatment lead to blend with a smoother
morphology a better intermixing between the polymer and the QDs and a more efficient exciton quenching effect.
References
[1] Günes S, Neugebauer, Sariciftci NS Chemical Reviews, 2007; 107(4): 1324−1338
[2] Nozik AJ Chem. Rev. , H 2000; 110: 6873-9.
[3] Huynh WU, Dittmer J J, Alivisatos AP Science 2002; 295:2425-8.
[4] S. Dayal and Burda Sl. Nano Letters 2010; 10: 239-45.
Figure 7: P3HT: QDs (Hexanoic Acid treat) films: 2D-height and phase AFM images a) and Section Analysis b)

Annalisa Bruno et al. / Energy Procedia 44 ( 2014 ) 167 – 175 175
[5] Greenham N, Peng X, Alivisatos AP. Physical Review B 1996; 54 17628
[6] Zhou Y , Riehle FS, Yuan Y, Schleiermacher HF, Niggeman M, Urban GA, Kruger M Appl. Phys. Lett.
2010; 96: 013304,-8;
[7] Lokteva I , Radychev N, Florian Witt , Holger Borchert , Jürgen Parisi and Joanna Kolny-Olesiak J. Phys.
Chem. C, 2010, 114 (29), 12784– 12791
[8] Celika D, Kruegerb M, Veitd C, Schleiermacherd H F, Zimmermannd B, Allarde S, Dumsche I, Scherfe U,
Rauschera F, Niyamakom P . Solar Energy Materials & Solar Cells 98 (2012), 433
[9] Scharber M.C, Mühlbacher D, Koppe M, Denk P, Waldauf C, Heeger AJ, Brabec CJ, Adv Material 2006;
18(6), 789-794
[10] Byoung-Ho Kang Tae-Yang You ; Se-Hyuk Yeom ; Kyu-Jin Kim ; Su-Hwan Kim ; Sang-Won Lee ; Heng
Yuan ; Dae-Hyuk Kwon ; Shin-Won Kang Photonics Technology Letters, IEEE 2012; 24(18):1594-1596
[11] Guilbert A Y, Reynolds L X, Bruno A, MacLachlan A, King S P, Faist M A, Pires E, Macdonald J E,
Stingelin N , Haque S A, Nelson J ACS Nano 2012; 6 (5): 3868–75.
[12] Singh S , Vardeny ZV Materials 2013; 6; 897-910.
[13] Parkinson P, Müller C, Stingelin N, Johnston MB, Herz L M J Phys Chem Lett, 2010; 1(19): 2788–2792
[14] de Villers BT, Tassone CJ, Tolbert SH, Schwartz BJ J Phys Chem C 2009; 113 (44): 18978-82
[15] Hoth CN, Choulis SA, P Schilinsky P, CJ Brabec CJ Adv Mater 2007; 19 3973-8