Solvent polarity and nanoscale morphology in bulk heterojunction organic solar cells:A case studyAjith Thomas, Anju Elsa Tom, Arun D. Rao, K. Arul Varman, K. Ranjith, R. Vinayakan, Praveen C. Ramamurthy,and V. V. Ison Citation: Journal of Applied Physics 115, 104302 (2014); doi: 10.1063/1.4867642 View online: http://dx.doi.org/10.1063/1.4867642 View Table of Contents: http://scitation.aip.org/content/aip/journal/jap/115/10?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Ferroelectric field effect of the bulk heterojunction in polymer solar cells Appl. Phys. Lett. 104, 253905 (2014); 10.1063/1.4885216 An analytical model for analyzing the current-voltage characteristics of bulk heterojunction organic solar cells J. Appl. Phys. 115, 034504 (2014); 10.1063/1.4861725 Optical anisotropy in solvent-modified poly(3,4-ethylenedioxythiophene):poly(styrenesulfonic acid) and its effecton the photovoltaic performance of crystalline silicon/organic heterojunction solar cells Appl. Phys. Lett. 102, 243902 (2013); 10.1063/1.4811355 Enhanced charge collection in confined bulk heterojunction organic solar cells Appl. Phys. Lett. 99, 163301 (2011); 10.1063/1.3651509 Open circuit voltage of stacked bulk heterojunction organic solar cells Appl. Phys. Lett. 88, 073514 (2006); 10.1063/1.2177633

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21

Solvent polarity and nanoscale morphology in bulk heterojunction organicsolar cells: A case study

Ajith Thomas,1,4,a) Anju Elsa Tom,1,a) Arun D. Rao,2 K. Arul Varman,2 K. Ranjith,2

R. Vinayakan,3 Praveen C. Ramamurthy,2,b) and V. V. Ison1,b)

1Centre for Nano-Bio-Polymer Science and Technology, Department of Physics, St. Thomas College, Pala,Kerala 686574, India2Department of Materials Engineering, Indian Institute of Science Bangalore, Karnataka 560012, India3Department of Chemistry, SVR NSS College Vazhoor, Kerala 686505, India4Research and Development Centre, Bharathiar University, Coimbatore, Tamilnadu 641046, India

(Received 26 December 2013; accepted 23 February 2014; published online 10 March 2014)

Organic bulk heterojunction solar cells were fabricated under identical experimental conditions,

except by varying the solvent polarity used for spin coating the active layer components and their

performance was evaluated systematically. Results showed that presence of nitrobenzene-

chlorobenzene composition governs the morphology of active layer formed, which is due to the

tuning of solvent polarity as well as the resulting solubility of the P3HT:PCBM blend. Trace

amount of nitrobenzene favoured the formation of better organised P3HT domains, as evident from

conductive AFM, tapping mode AFM and surface, and cross-sectional SEM analysis. The higher

interfacial surface area thus generated produced cells with high efficiency. But, an increase in the

nitrobenzene composition leads to a decrease in cell performance, which is due to the formation of

an active layer with larger size polymer domain networks with poor charge separation possibility.VC 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4867642]

I. INTRODUCTION

Solar cells based on organic semiconducting polymers

have received considerable attention in the past few years due

to their potential of providing light weight, flexible, environ-

mentally safe, and inexpensive solar cells using solution proc-

essing methods.1 Due to their strong absorption cross-section,

they are also ideal for fabricating thin film based devices,

which makes them efficient in materials saving. A commonly

used organic solar cell configuration employs an active layer

formed out of an electron donor-acceptor heterojunction,

comprising a conjugate polymer and a fullerene derivative, to

separate the photo-generated excitons. Since the conjugated

polymers have strong electron donating properties and the

fullerene derivatives are electron acceptors, the charge sepa-

ration in organic solar cells occurs by photo induced electron

transfer between the two components.2,3 But, a simple bilayer

interface is insufficient to serve the purpose effectively due to

the short exciton diffusion length of polymers by virtue of

their lower dielectric constant values. A strategy that is

adopted to overcome this issue is the bulk-heterojunction

(BHJ) concept, in which the donor-acceptor phases are mixed

to form a three dimensional interpenetrating network (distrib-

uted donor-acceptor junction) throughout the active region, to

separate the excitons. The distributed active layer serves not

only as the interface for charge separation within the diffu-

sion range but also as the percolation pathways for efficient

charge carrier transport to the respective electrodes.4 An

active layer of sufficient thickness with a nanoscale morphol-

ogy, ensuring a balance between large interface area and

continuous pathways for efficient charge transport, deter-

mines the power conversion efficiency of the organic solar

cells. Therefore, the research on organic solar cells requires

more attention in tuning the active layer morphology so as to

contribute to high efficiency.

The present study is focused on analyzing the nano-

scale morphology and the resulting cell performance of

poly(3-hexylthiophene) (P3HT)-[6,6]-phenyl C61-butyric

acid methyl ester (PCBM) based solar cells subjected to

certain pre/postdeposition treatments. Though P3HT:PCBM

solar cells have reached a power conversion efficiency of

about 9%, there are still certain intrinsic issues to be addressed

for further improvement.5 Conventionally, the polymer and

the fullerene components are spin coated together using a

common solvent and as the solvent gets evaporated, a hetero-

junction network spanning the entire active layer comprising

phase separated polymer and fullerene domains is formed and

it has been reported that the morphology of the active layer

formed depends strongly on the processing conditions.4–7

Several approaches have been adopted by different research

groups in this direction that include thermal annealing of the

active layer, solvent annealing, usage of suitable additives to

the coating solvent, etc. After the thermal annealing, the

P3HT:PCBM blend was found to adopt a well organized

structure resulting in a better crystalline phase, a condition

favorable for efficient charge transport.8,9 The solvent anneal-

ing enhances the cell performance by allowing the active layer

components to remain partially dissolved for longer time peri-

ods so that their diffusion occurs at a higher rate resulting in

an improvement of the P3HT crystallinity.10–13

The presence of a processing additive is also shown as

an effective route to control the bulk heterojunction mor-

phology. It has been reported that the addition of a miscible

a)A. Thomas and A. Elsa Tom contributed equally to this work.b)Authors to whom the correspondence should be addressed. Electronic

addresses: isonvv@yahoo.in and praveen@materials.iisc.ernet.in

0021-8979/2014/115(10)/104302/5/$30.00 VC 2014 AIP Publishing LLC115, 104302-1

JOURNAL OF APPLIED PHYSICS 115, 104302 (2014)

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21

dipolar solvent nitrobenzene (NB) to the coating solution

chlorobenzene (CB) results in a photo-conversion efficiency

of about 4% in P3HT:PCBM based solar cells.14–16 Here, the

observed effect of the dipolar solvent additive on cell per-

formance is explained based on the crystallinity analysis

from optical absorption studies and the topography analysis.

In our study, we have utilized the conductive-Atomic Force

Microscopy (c-AFM) technique, which is more reliable than

the conventional phase imaging to distinguish the polymer

and the fullerene domains. The active layer morphology was

correlated with the solvent polarity by systematically varying

the solvent composition. Also, we have resorted to cross-

sectional and surface Scanning Electron Microscopy (SEM)

to achieve a better view of the final active layer morphology.

The variations in the cell performance with the solvent com-

position have been discussed in terms of the charge separa-

tion and transport capabilities of the distributed

heterojunction domains. To the best of our knowledge, so

far, a systematic study on the effect of solvent polarity in

governing the nanoscale morphology has not been performed

on this system. The performance evaluation of a set of cell

structures grown under different NB-CB compositions can

give an insight to the active layer growth phenomenon.

II. EXPERIMENTAL

A. Solar cell fabrication

The cell structures were fabricated in the traditional con-

figuration ITO/PEDOT:PSS/P3HT:PCBM/Al. The etched

ITO coated glass plates were processed following standard

procedures.10 The hole transporter layer PEDOT:PSS

(poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate))

was filtered through a 0.45 lm filter and spin coated over the

ITO substrates with a spin speed of 3000 rpm followed by

another spin speed of 1000 rpm, both for 30 s, resulting in a

layer thickness of about 40 nm. The substrates were then sub-

jected to a thermal annealing for 3 min at 130 �C. P3HT

(15 mg/ml) and PCBM (15 mg/ml) in the ratio 1:1 were dis-

solved in CB and stirred overnight in an inert atmosphere.

NB was added to the coating solution 2 h before the active

layer spin coating. The coating solution was then filtered

through a 0.45 lm filter and spin coated over the

PEDOT:PSS layer at a spin speed of 1000 rpm. The average

thickness of the active layer formed was �250 nm and

finally, Al electrodes were deposited (�100 nm) over the

active layer by thermal evaporation, under a pressure

�10–5 Pa.

B. Characterization

C-AFM and tapping mode AFM were performed using a

Bruker Dimension Icon Atomic Force Microscope. A Karl

Zeiss Ultra 55 Field Emission Scanning Electron

Microscope was used for surface and cross-sectional analy-

sis. The solar cell characterizations were done using an Oriel

Sol3A Class AAA solar simulator under standard AM 1.5G

illumination. A Perkin Elmer Lambda 35 UV-Visible spec-

trophotometer was used for electronic absorption measure-

ments. For solar cell fabrication, ITO, P3HT, PCBM, and

PEDOT:PSS were procured from Kintec, Rieke Metals,

Nano-C (99.5%), and Sigma Aldrich, respectively. Reagents

and solvents were purchased from Merck and further purifi-

cations were done wherever required.

III. RESULTS AND DISCUSSION

Five sets of ITO/PEDOT:PSS/P3HT:PCBM/Al based

heterojunction solar cells were fabricated under identical

conditions, except in the composition of the active layer

coating solvent. Different NB-CB combinations (with NB

composition varying from 0%–4% (v/v)) for dissolving

P3HT and PCBM prior to spin coating were used and its

effect on the cell performance was evaluated. The results

obtained with 0, 0.5, 1, 2, and 4% of NB (labeled as CB,

NB1, NB2, NB3, and NB4, respectively) are tabulated in

Table I. In addition, these results were also confirmed by

repeating the studies with the same samples, after annealing

at 140 �C (labeled as CBa, NB1a, NB2a, NB3a, and NB4a,

respectively). The corresponding J-V characteristics are

shown in Fig. 1.

Comparison of the cell performance in presence and ab-

sence of NB clearly indicates that the additive indirectly

plays a crucial role in the charge generation and separation

FIG. 1. J-V Characteristics of devices, before and after annealing. (A few

insignificant plots are omitted for clarity.)

TABLE I. Performance evaluation of solar cells fabricated by varying NB-

CB composition (“a” denotes annealed sample. The sample NB4a showed

poor efficiency).

Devices Voc (V) Jsc (mA/cm2) FF (%) Efficiency (%)

CB 0.71 2.67 31.84 0.61

CBa 0.58 9.7 43.74 2.46

NB1 0.60 7.61 30.33 1.38

NB1a 0.60 10.65 47.58 3.04

NB2 0.55 7.0 33.08 1.27

NB2a 0.55 7.31 37.21 1.50

NB3 0.54 6.21 34.97 1.17

NB3a 0.54 5.76 35.45 1.10

NB4 0.53 2.34 33.07 0.41

NB4a 0.49 … … …

104302-2 Thomas et al. J. Appl. Phys. 115, 104302 (2014)

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21

mechanism and influences the overall cell performance. A

trace amount (0.5%) of NB contributed a �200% increase in

efficiency of the cell when compared to that fabricated solely

out of CB (labeled as NB1 and CB, respectively). This obser-

vation was supported by a similar enhancement in cell per-

formance when the corresponding annealed samples were

used for device fabrication (labeled as NB1a and CBa,

respectively).

The scenario changed when the cells were fabricated

with higher compositions of NB. When a 1% (v/v) NB-CB

mixture (labeled as NB2) was used for spin coating the

active layer components, it reversely affected the cell per-

formance, i.e., the overall efficiency was decreased. This is

well supported by the results obtained for the corresponding

annealed sample (labeled as NB2a). Further, the studies con-

ducted with higher compositions of NB (2 and 4%) also

showed a similar trend for both the as grown and the

annealed samples.

The results underline that the cell parameters, such as

Jsc, Voc, and FF, are governed by the NB-CB composition. A

reduction in Voc observed in presence of NB arises from an

increase in crystallinity favouring the charge carrier mobil-

ity.17,18 The poor performance of the device CB is mainly

due to fragmental inter-penetrating networks formed and

also due to poor crystallinity, which is evident from Fig. 2.19

The improved performance of NB1 compared to CB can

be justified by the three fold increase in the current density

(Jsc), which is a clear indication of the efficient charge sepa-

ration in the device. This is attributed to the increased inter-

facial area of the P3HT:PCBM domains formed in the active

layer. Obviously, the presence of NB causes the formation of

aggregated/crystalline P3HT networks during the drying pro-

cess resulting in an improved device performance.15 It indi-

cates that in NB1, the presence of NB favours the formation

of domains having size in the order of diffusion length of

charge carriers to effect good charge separation and trans-

port. Though NB1 showed high Jsc, it suffered inferior FF,

which can be due to the presence of the high boiling point

NB residue in the device. After spin coating, the device NB1

was not observed perfectly dry indicating the presence of the

residual NB. To validate this, we have compared the FF of

the device NB1 with a sample of the same annealed at 70 �C(data not provided in Table I). It is observed that after

annealing, the FF got improved (30.33%–43.32%), due to

the removal of the residual NB. Again, the observed change

in FF in case of NB1a is comparable (30.33%–47.58%) to

that of at 70 �C. As per previous studies, for systems not

involving NB, annealing at 70 �C contributed only nominally

to the FF enhancement.20 Note that the improved perform-

ance of the corresponding annealed sample (NB1a) is due to

the further increased crystallinity of the polymer domains

during high temperature processing.

These factors were examined with the help of c-AFM,

tapping mode AFM and SEM. C-AFM was carried out by

mapping the current image over the surface at a fixed (posi-

tive) bias to identify the polymer networks. The hole current

was extracted using a conducting tip (MESP-Bruker) and the

image contrast represents the differences in the hole conduc-

tivities of the materials forming the blend. The AFM results

showing the topography and the current distribution for the

device NB1a (obtained from the same area) are presented in

Fig. 3.

The topography image exhibits a higher peak to valley

height and roughness. The c-AFM image shows that the

P3HT domains are of bigger size. The images also show that

there exists a correlation between the height and the current

signals in the whole area. To confirm this, we have plotted

the height and the current signals against a common direc-

tion, which is shown in Fig. 4. The one to one correspon-

dence observed between the height and current variations

FIG. 2. XRD pattern of the devices CB and NB1.

FIG. 3. (a) Topographic image and (b) current image of the device NB1a.

FIG. 4. Height/current signals against a common direction in the AFM/

c-AFM images of sample NB1a.

104302-3 Thomas et al. J. Appl. Phys. 115, 104302 (2014)

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21

confirms that the hill areas in the topographic image corre-

spond to the P3HT domains.

Apparently, the active layer morphology in the bulk was

found to be different from that observed in the surface,

favouring efficient charge separation and transport leading to

a high efficiency in the case of sample processed with 0.5%

NB composition. This is evident from the cross-sectional

SEM image of the sample, which is shown in Fig. 5. The

image shows that the larger domains are confined to the

active layer surface only and smaller P3HT domains fill the

bulk so that there exists enough interfacial area for charge

separation which is evident from the high Jsc value obtained

for NB1 and NB1a (Table I). In the SEM image, the brighter

regions on the surface are P3HT domains.21 It may also be

noted that the size of the P3HT domains in the surface

obtained from AFM is matching with that obtained from the

cross sectional SEM image, which is a clear indication of the

fact that the peak to valley height is a direct representation of

the polymer domains thickness.

All these observations confirmed that the presence of

smaller P3HT domains in the bulk active layer formed while

processing with trace amount of NB (0.5%) contributes to an

efficient cell performance.

But higher NB compositions (samples NB2-NB4) lead

to an unfavourable active layer morphology as evident from

the decreased cell performance. This is attributed to the for-

mation of P3HT domains with larger dimensions, resulting

in less efficient charge separation. To evaluate this aspect,

the morphology of annealed sample NB4a, sample with

highest composition of NB, was analysed using AFM and

surface SEM and the images obtained are shown in Fig. 6.

The surface topography (Fig. 6(a)) showed a peak to

valley height of about 300 nm, which is comparable with the

active layer thickness (Fig. 5). This means that the P3HT

domains thickness extends in the entire depth of the active

layer, reducing the polymer-fullerene interfacial area for

effective charge separation, which adversely affects the cell

performance. This assumption is further supported by the

SEM image, shown in Fig. 6(b), exhibiting a large PCBM

agglomerate and the surrounding P3HT aggregated net-

works. It may therefore be concluded that the poor cell

performance in NB4a is due to the unfavourable morphology

for charge separation.

The effect of NB on the active layer morphology can be

explained based on the solvent polarity and solubility of

P3HT and PCBM in NB-CB solvent. Note that P3HT is less

soluble (15.9 mg/ml) in CB than PCBM (59.5 mg/ml).14,22

Since NB is a non-solvent for P3HT and is a poor solvent for

PCBM, the solubility of both P3HT and PCBM decreases

with the addition of NB. At low NB composition (in case of

NB1), the resulting polarity and solubility of solvent favours

formation of small sized P3HT and PCBM networks during

spin coating. As the active layer spin coating proceeds, the

NB to CB concentration ratio increases due to the slow evap-

oration of the high temperature boiling NB (210.9 �C) com-

pared to CB having a boiling point of 131 �C. The result is a

faster reduction of the P3HT solubility compared to PCBM

so that the P3HT component reaches a level of super-

saturation initially, forming larger crystalline domains upon

further evaporation of the solvent. PCBM stays in solution

for longer time periods before it reaches saturation and

finally they precipitate simultaneously resulting in an inter-

penetrating donor-acceptor network. It is reasonable to think

that the rate of precipitation of P3HT is higher than that of

PCBM. The enhanced performance of the sample NB1 in

comparison with CB arises due to a better-organized mor-

phology achieved in them by the above mechanism.

As the NB concentration is increased, the difference

between the saturation points of P3HT and PCBM increases

to a larger extent so that the P3HT aggregates consume more

and more of the active layer thickness. For the NB concen-

tration 0.5% (v/v), the difference between the saturation

points is relatively small so that the PCBM component starts

precipitating just after P3HT begins to aggregate. Therefore,

the thickness of the polymer aggregate is limited only to a

portion of the active layer thickness and the rest of film has

partially crystalline/amorphous P3HT networks. The thermal

annealing improves the crystallinity of these partially crys-

talline/amorphous regions. An appreciable change of absorb-

ance of the device NB1 can be seen after it is annealed

(Fig. 7). The strengthened absorption after annealing is

attributed to the further crystallization of P3HT.23 Higher

NB concentrations make the device less responsive to ther-

mal annealing, which is evident from the absorption spectra

of NB2 and NB2a. The effect arises because the amount of

amorphous/partially crystalline P3HT remains in the film

FIG. 5. Cross sectional SEM image of the sample NB1a.

FIG. 6. (a) Tapping mode AFM image and (b) surface SEM image of the

sample NB4a.

104302-4 Thomas et al. J. Appl. Phys. 115, 104302 (2014)

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21

after the active layer coating is very low and most of the

P3HT is in the form of larger aggregates.

The polarity of solvent and in turn the solubility of poly-

mer is changed while increasing the composition of NB in

NB-CB solvent. With higher percentage of NB, larger sized

polymer domains extending entire active layer is resulted

during spin coating process, which is unfavourable for effec-

tive charge separation.

IV. CONCLUSIONS

Our study showed that solvent polarity controls the sol-

ubility of the P3HT:PCBM blend and the resulting active

layer morphology in organic bulk heterojunction solar cells.

C-AFM, tapping mode AFM, and surface and cross-

sectional SEM analysis confirmed that lower amount of NB

favoured the formation of P3HT domains with higher inter-

facial surface area which generated cells with better per-

formance. But, an increase in the NB composition leads to

the formation of active layer with higher size polymer do-

main networks with poor charge separation possibility and

in turn poor cell efficiency. A detailed study with various

solvents and solvent compositions may give more insight

into the morphology evolution in bulk heterojunction solar

cells.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the INUP facility of

Centre for Nanoscience and Engineering (CeNSE), Indian

Institute of Science (IISc), Bangalore for solar cell fabrica-

tion and performance evaluation, AFM and SEM analysis.

Author A.E.T. thank DST, Govt. of India, for providing fel-

lowship (Order No. SR/FTP/PS-108/2010).

1A. Manor, E. A. Katz, T. Tromholt, B. Hirsch, and F. C. Krebs, J. Appl.

Phys. 109, 074508 (2011).2F. Padinger, R. S. Rittberger, and N. S. Sariciftci, Adv. Funct. Mater. 13,

85 (2003).3P. Schilinsky, C. Waldauf, and C. J. Brabec, Appl. Phys. Lett. 81, 3885

(2002).4H. Hoppe and N. S. Sariciftci, J. Mater. Chem. 16, 45 (2006).5L. Dou, J. You, J. Yang, C.-C. Chen, Y. He, S. Murase, T. Moriarty, K.

Emery, G. Li, and Y. Yang, Nature Photon. 6, 180 (2012).6P. Kumar, S. C. Jain, V. Kumar, S. Chand, and R. P. Tandon, J. Appl.

Phys. 105, 104507 (2009).7X. Yang and J. Loos, Macromolecules 40, 1353 (2007).8X. Yang, J. Loos, S. C. Veenstra, W. J. H. Verhees, M. M. Wienk, J. M.

Kroon, M. A. J. Michels, and R. A. J. Janssen, Nano Lett. 5, 579

(2005).9U. Zhokhavets, T. Erb, H. Hoppe, G. Gobsch, and N. S. Sariciftci, Thin

Solid Films 496, 679 (2006).10G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, and Y.

Yang, Nat. Mater. 4, 864 (2005).11G. Li, Y. Yao, H. Yang, V. Shrotriya, G. Yang, and Y. Yang, Adv. Funct.

Mater. 17, 1636 (2007).12V. Shrotriya, Y. Yao, G. Li, and Y. Yang, Appl. Phys. Lett. 89, 063505

(2006).13T. Erb, U. Zhokhavets, G. Gobsch, S. Raleva, B. Stuhn, P. Schilinsky, C.

Waldauf, and C. J. Brabec, Adv. Funct. Mater. 15, 1193 (2005).14L. Chang, H. W. A. Lademann, J.-B. Bonekamp, K. Meerholz, and A. J.

Moule, Adv. Funct. Mater. 21, 1779 (2011).15A. J. Moule and K. Meerholz, Adv. Mater. 20, 240 (2008).16G. Dennler, M. C. Scharber, and C. J. Brabec, Adv. Mater. 21, 1323 (2009).17M. M. Mandoc, L. J. A. Koster, and P. W. M. Blom, Appl. Phys. Lett. 90,

133504 (2007).18J.-T. Shieh, C.-H. Liu, H.-F. Meng, S.-R. Tseng, Y.-C. Chao, and S.-F.

Horng, J. Appl. Phys. 107, 084503 (2010).19H.-Y. Chen, H. Yang, G. Yang, S. Sista, R. Zadoyan, G. Li, and Y. Yang,

J. Phys. Chem. C 113, 7946 (2009).20W. Ma, C. Yang, X. Gong, K. Lee, and A. J. Heeger, Adv. Funct. Mater.

15, 1617 (2005).21H. Hoppe, M. Niggemann, C. Winder, J. Kraut, R. Hiesgen, A. Hinsch, D.

Meissner, and N. S. Sariciftci, Adv. Funct. Mater. 14, 1005 (2004).22F. Machui, S. Langner, X. Zhu, S. Abbott, and C. J. Brabec, Sol. Energy

Mater. Sol. Cells 100, 138 (2012).23M. T. Dang, G. Wantz, H. Bejbouji, M. Urien, O. J. Dautel, L. Vignau,

and L. Hirsch, Sol. Energy Mater. Sol. Cells 95, 3408 (2011).

FIG. 7. The P3HT absorption spectra of samples NB1 and NB2, before and

after annealing.

104302-5 Thomas et al. J. Appl. Phys. 115, 104302 (2014)

[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:

203.200.35.19 On: Wed, 29 Apr 2015 11:16:21