Solar Energy Materials & Solar Cells 95 (2011) 1326–1332

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

0927-02

doi:10.1

n Corr

Departa

Murcia,

E-m

journal homepage: www.elsevier.com/locate/solmat

The influence of UV radiation and ozone exposure on the electronic propertiesof poly-3-octyl-thiophene thin films

Jose Abad a,b,n, Nieves Espinosa c, Rafael Garcıa-Valverde c, Jaime Colchero a, Antonio Urbina c,d

a Centro de Investigacion en Optica y Nanofısica, Departamento de Fısica, Campus Espinardo, Universidad de Murcia, E-30100 Murcia, Spainb Departamento de Fısica Aplicada, Universidad Politecnica de Cartagena, Doctor Fleming, s/n, 30202 Cartagena, Spainc Departamento de Electronica, Tecnologıa de Computadoras y Proyectos, Universidad Politecnica de Cartagena, Plaza del Hospital s/n, 30202 Cartagena, Spaind Department of Physics, Blackett Laboratory, Imperial College London, Prince Consort Road, London SW7 2BW, United Kingdom

a r t i c l e i n f o

Article history:

Received 14 November 2010

Received in revised form

9 December 2010

Accepted 12 December 2010Available online 6 January 2011

Keywords:

Organic solar cells

Poly-alkyl-thiophenes

Scanning force microscopy

Electrostatic scanning force microscopy

Electronic transport

UV-degradation

48/$ - see front matter & 2010 Elsevier B.V. A

016/j.solmat.2010.12.032

esponding author at: Centro de Investigac

mento de Fısica, Campus Espinardo, Unive

Spain. Tel.: +34 868 07 1096; fax: +34 968 3

ail address: jose.abad@upct.es (J. Abad).

a b s t r a c t

The modification of the conductivity as well as the nanoscale morphology and electrostatic properties of

poly-3-octyl-thiophene (P3OT) thin films have been studied as these films are irradiated by ultraviolet

light (UV). Films of about 100 nm thickness were prepared by spin-coating a P3OT solution in toluene on

glass substrates. The samples were characterized by electronic transport measurements and non-contact

scanning force microscopy (NC-SFM) for every cycle of UV radiation. Nanoscale topographic and

electrostatic characterization using NC-SFM were performed on the same location using a specially

designed sample holder. A two stage degradation process has been observed, the first one presents a

chemical modification of the polymer (decoloration of sample) and free carrier mobility reduction, the

second one is characterized by a strong structural modification, thickness reduction, oxygen doping and

further mobility reduction. The comparison of these results on P3OT with other studies performed on

P3HT will allow for a detailed analysis of the role of side-chains in the degradation mechanism upon UV

irradiation.

& 2010 Elsevier B.V. All rights reserved.

1. Introduction

Organic solar cells utilize a blend of two conjugated polymers ora conjugated polymer and a fullerene derivative as active layer forefficient solar energy conversion [1]. Apart from the active layer,several polymeric thin films are included in the device in order toimprove charge extraction from the active layer towards theelectrodes. The most successful class of polymers used as donormaterial (p-type material in a typical semiconducting junction) arethe poly(3-alkylthiophenes) (PAT), which have been intensivelystudied due to their good environmental stability, favorableprocessability, and easy modification of optical and electronicproperties [2–7]. In the field of organic solar cells the powerconversion efficiency and the stability are issues that have to beaddressed before this technology can compete with traditionalsilicon solar cells. The degradation mechanisms of the organic solarcells are now being intensively studied by several groups [8–10].Many causes of degradation can be roughly organized into twobroad groups: a first class of external agents affecting the deviceperformance (contamination by oxygen, water and UV irradiation

ll rights reserved.

ion en Optica y Nanofısica,

rsidad de Murcia, E-30100

2 5337.

have been pointed out as the main causes of degradation) andwhich can be solved by improved encapsulation and UV filtering,and a second class which are structural or chemical processesoccurring in the polymers (bulk heterojunction nanostructureevolution, embedded interlayer roughness modification, diffusionof species between interfaces of different layers, metal contamina-tion from the electrodes, etc.). This second class of processes can betriggered by an external agent, UV irradiation being the mostaggressive one, but all are dynamic processes inherently occurringat room temperature in a polymeric device, which does not have astable configuration at the time scale of months or years. A deeperunderstanding of the degradation mechanisms of poly-alkyl-thio-phenes upon UV irradiation is strongly needed, and accordingly,the means to improve the poor UV light stability and device lifetimeis an important field of research [9–11]. The structure of thedegraded polymer must be examined for a deeper understanding ofthe degradation mechanisms which is necessary for the design oflong lifetime and UV stable devices. In this article we will focus onpoly-3-octyl-thiophenes because the comparison of the behavior oftwo polymers which only differ in the size of the side chain willshed some light on the proposed mechanism of side chaindegradation. Furthermore, solar cells in which P3OT and functio-nalized single wall carbon nanotubes were proposed, some timeago, as very stable (but not so efficient) active layers for organicsolar cells [14,15].

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–1332 1327

The main degradation mechanism that we have studied is UVirradiation, but since all the experiments have been performed inair, also degradation of the polymeric films by the generated ozonehas to be taken into account. Although ozone is well-known as adegrading agent in polymer chemistry [16–19] and the effect ofdoping and de-doping processes on the conductivity of P3OT wasreported long time ago [20,21], systematic studies on the interac-tion of oxygen with poly-alkyl-thiophenes in terms of doping anddegradation have appeared more recently. The effect of oxygenexposure has been found to have a clear effect in the polymerconductivity, which has been explained by the formation ofreversible charge transfer complexes of the oxygen with thepolymer which facilitates the generation of carriers upon applica-tion of an electric field [22–24]. This results in a p-doping of thepolymer, this doping and de-doping process have been found fullyreversible in the absence of light [25]. Some authors link thischange in the electronic structure of the polymer with morpholo-gical changes during the degradation process [10,26,27], but in thiscase, the changes are no longer reversible. Some mechanismswhich explain the irreversible degradation in P3HT point towardsmodification of side chains of the polymer as one of the mainmechanisms of degradation of the active layer [8,12,13]. The effectof side chain length has also been proposed as a mechanism, whichexplains the different density of states due to differences ineffective conjugation length and measured in samples of P3OTsandwiched between metal plates [28]. A different approach to theexplanation of the formation of charge transfer complexes uses theconcept of trap states created by the inclusion of oxygen in the bulkpolymer, these studies use thermally stimulated current andcharge extraction by linearly increasing voltage to experimentallymeasure the intrinsic density of trapped states, which are found tohave Gaussian energy distributions with different centers [29,30].A detailed mechanism has been proposed theoretically to explainthe formation of the charge transfer complexes via the hybridiza-tion of O2 and polymer orbitals; in this case, the Fermi level ispushed down into the valence band and pinned with the oxygenband due to charge transfer to O2, therefore explaining a relation-ship between the amount of transferred electrons and the ioniza-tion potential of the host materials [31,32]. This ionizationpotential should better be considered as an ionization energy inthe case of conjugated polymers, where the HOMO level is a valenceband extended to the conjugation length of the polymer andtherefore, polymers with a lower HOMO will be more prone tothis oxygen p-doping effect.

All these changes may result in an improvement of conductivityby oxygen doping if post-processing treatments are carefullytailored in order to improve the performance of devices in termsof better filling factor and elimination of inflexion points in the IVcurves [33].

In this article we have studied thin films of P3OT by means ofnon-contact dynamic surface force microscopy and electronictransport measurements. The combination of probes which allowsmeasurements at both the nanoscale and the macroscale enables toinvestigate correlation between the modification of the nanos-tructure and its impact on macroscopic electronic properties. Theelectrostatic behavior of P3OT thin film samples by means ofelectrostatic force microscopy (ESFM) on a nanometer scale and itsrelationship with the mobility change observed upon UV irradia-tion are presented and discussed.

2. Experimental

Regioregular P3OT was purchased from Sigma-Aldrich, with98.5% head to tail couplings, molecular weight Mn¼54,000, poly-dispersity index D¼2.6 and density r¼1.05 g/cm3 [34]. With the

molecular weight of mMol¼194 g/Mol for a P3OT monomer(C12H18S) a number density of nP3OT¼r/mMolE3.26�10�27 par-ticles/m3 is obtained, corresponding to a volume of 0.67 nm3 permonomer. P3OT thin films of approximately 110 nm thicknesswere prepared on glass cover slips by spin-coating a 10 mg/mlsolution of P3OT in toluene at 2500 rpm for 90 s in air at roomtemperature.

UV/O3 exposure was performed using a commercial 50 WUV/ozone cleaning system from Novascan Technologies (Ames,IA, USA). This system delivers most of its energy (about 50%) intothe 254 nm peak and 5% into the more energetic 185 nm peak. Forour experimental setup, the samples are illuminated homoge-neously with an irradiance of 2.5 kW/m2 for the main wavelengthof l0¼254 nm, which is about 320 times more UV energy thanthat delivered by sunlight at AM1.5G standard conditions. Thecorresponding photon flux is jph¼ I0/(hn0)¼ I0l/(hc)E3.2�1021

photons/m2/s.The experimental setup for the electronic transport measure-

ments is a two probe system, where the current is measured withan electrometer (Keithley Mod. 6514) while a controlled voltage isapplied to the sample with a programmable voltage source whichacts as an active load (Keithley Mod. 220). Tungsten probes withcontrolled contact pressure (Suss Microtech probes) have beenused together with silver paint in order to improve the contact. Thegeometry of the samples is bar-shaped, where stripes of P3OT areleft after cutting a rectangular P3OT shape on the glass substrates.The silver paint is applied at both ends of the stripes. Thisconfiguration allows us to obtain accurate plots of current intensityversus applied voltage, and therefore we can easily obtain aresistance value from linear fits provided that the resistancebehaves linearly (ohmic contacts). We used triaxial cables fromthe electronic equipment to the probes, but the final contact ismade using a coaxial cable and the final tungsten tip, therefore, theguard does not contact the probe. Since we are measuringresistance values of the order of mega Ohms, the small voltagedrop along the final section of the coaxial and the probe isnegligible. All the measurements were performed in air and atroom temperature.

A Nanotec Electronica SFM system with a phase locked loop(PLL)/dynamic measurement board [35] was used with OlympusOMCL-AC-type Si cantilevers (nominal force constant: 2 N/m;resonance frequency: 70 kHz). Imaging was performed in non-contact dynamic SFM using the oscillation amplitude as feedbackchannel to maintain a constant tip-sample interaction, and a smallreduction of oscillation amplitude (AsetE0.95 Afree) was chosen forfeedback. A new precision sample holder integrated in the NanotecSFM system has been developed for the experiments presented inthis work. Essentially, this sample holder allows ex-situ manipula-tion of the sample and SFM imaging of the same region with a veryaccurate re-positioning. We can therefore attribute changes ofmorphology or other properties to a particular effect of UVradiation and/or ozone.

An external lock-in was employed for Electrostatic ScanningForce Microscopy (ESFM) measurements. For the experimentsreported in the present work, the tip was biased with a DC voltage(VDC) of 1 V and an AC voltage (Vac) of 0.5 V at an electricalmodulation frequency nelec¼7 kHz. The phase shift signal of themechanical oscillation (measured by the dynamic measurementboard) induced by this electrical modulation frequency is the inputfor the external lock-in used for the electrical measurements [36].Further details of the ESFM setup and working operation modes aredescribed elsewhere [37,38]. Briefly, we recall that in the (true)non-contact regime, the electrostatic interaction is Ielec(d,Vtip)¼�1/2C(d)(Vtip�VCP)2, where d is the tip-sample distance, Vtip the tipvoltage, C(d) the capacitance of the tip-sample system and VCP the(local) contact potential between tip and sample. For low

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–13321328

oscillation amplitudes, the resonance frequency shift induced bythe electrostatic interaction has the same voltage dependence, butwith higher derivatives of the capacitance:

Dnd_ffin0Ielec

00 ðdÞ=ð2cleverÞffin0C00ðdÞðVbias�VCPÞ2=ð4cleverÞ

where C00(d) is the second derivate of the capacitance, n0 the naturalfrequency and clever the spring constant of the cantilever. Thisfrequency shift will induce a variation of the phase of themechanical oscillation at the frequency nelec, which is proportionalto the local capacitance of the sample (more precisely: C00(d)) andthe contact potential difference Vbias–VCP between tip and sample.

3. Results and discussion

3.1. Surface morphology

Fig. 1 shows a series of medium scale images (20�20 mm2)taken for the pristine sample (A) and for different times ofillumination (B–H), ranging from 1 min to a maximum of 30 minof total irradiation. We note that these images show the sameregion of the sample, which is only possible using the special setup

Fig. 1. A series of SFM topographic images (20�20 mm2) showing the morphology of th

(F) 10 min, (G) 15 min and (H) 30 min of UV irradiation. In all images Dz¼50 nm. (i) RM

outlined the area taken for the RMS roughness analysis.

for the sample holder described above. The sample is thus imaged,then I–V measured and irradiated ex-situ and then preciselyre-allocated in the SFM system in order to resume the imagingexperiment at exactly the same spot of the sample. A detaileddescription of this kind of experiments and a discussion of how theUV irradiation varies the morphology of the P3OT thin films will bepublished elsewhere [39]. Fig. 1A shows a typical image of apristine P3OT thin film. The surface is generally quite flat, althoughlocally some bright features significantly increase the overallsurface roughness: local surface roughness for 2�2 mm2 flatregions is 1.270.2 nm vs. 3.670.8 nm (small features) to100 nm (biggest feature) for regions with the bright features. Inaddition, most of these features have a relatively well definedlateral dimension of 300–500 nm. We think that these roundstructures could be polymer aggregates which have not been welldissolved during the preparation method or small gas bubblestrapped in the region between the polymer and the substratesurface. In the first 5 min of exposure no significant changes on themorphology of the samples are observed (Fig. 1B–D). Interestingly,some of the larger bright structures develop holes in their center(marked in the image with an II) with a depth between 10 and20 nm. Jørgensen et al. [10] and Norrman [27] have reported the

e P3OT thin film (A) pristine sample, (B) after 1 min, (C) 3 min, (D) 5 min, (E) 7 min,

S roughness, from all the images of Fig. 1, as function of the UV irradiation. In (A) is

Table 1Evolution of the mean thickness as a function of UV irradiation. The error in the

determination of film mean thickness is typically 73 nm.

Irradiation (min) 0 1 3 5 7 10 15 30 45 60

Thickness (nm) 108 104 105 108 104 97 90 60 37 25

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–1332 1329

presence of microscopic holes, which were related to the formationof pinholes during solar cell fabrication. These microscopic holesshould be channels for the oxygen and water entrance. However,our results could indicate that the formation of these pinholeshappens once the device is exposed to UV light irradiation andozone exposure after the device fabrication. After 7 min also thelargest structure (lower left part of the image) starts to develop ahole in its center. In addition, around this structure a higher rim(4 nm) develops, which becomes more pronounced (8 nm ofheight) after 10 min of UV irradiation (Fig. 1F).

The most important variations of the surface morphology areobserved after 15 min of irradiation (Fig. 1G), where small grain-like structures appear over the whole surface. Moreover, for thecase of the largest structure, the hole reaches all the way down tothe glass substrate (depth about 100 nm) and a characteristic ringof small spherical particles-satellites- is formed around the largerstructures that were present from the beginning. The regionbetween the initial central particle and the satellites has a lowerroughness than the rest of the surface. Finally after 30 min (Fig. 2F),the small grain-like structures coalesce to form larger featuresmaking the surface more blurred and the overall RMS roughnesshas increased significantly (16 nm), as shown in the graph 1 I). Inorder to calculate the evolution of RMS roughness with the UVirradiation time a rectangle of 10�13 mm2 centered on the featuremarked I (Fig. 1A) has been analyzed for each image. Within theexperimental error the roughness is almost constant during thefirst 10 min of UV irradiation and increases significantly after10 min. Therefore in view of the Fig. 1, the main morphologychanges are induced between 10 and 15 min of UV irradiation.

To estimate the height of the films, they were very carefullyscratched off in order to expose the substrate. The correspondingtrench was imaged by DSFM to determine the film thickness aftereach cycle of UV irradiation in order to study the relationshipbetween UV irradiation time and film thickness. In particular wewere interested to determine whether the UV irradiation induced

Fig. 2. A series of ESFM electrostatic images (20�20 mm2) acquired simultaneously with

5 min, (C) 7 min, (D) 10 min, (E) 15 min, (F) 30 min of UV irradiation. In all images Dz¼

removal of material from the film, or only as just discussed- anincrease of surface roughness. The results are shown in Table 1. Wehave found that during the first 10 min of irradiation the filmthickness stays almost constant around 100 nm. However, after10 min the thickness starts to decrease approximately linearly withthe exposure time until a thickness of 25 nm is reached for 60 minof irradiation. It is remarkable that the decrease of film thicknessafter 10 min of irradiation is coincident with the main changesobserved in the morphology of the samples, and also with theincrease of surface roughness.

3.2. Electrostatic properties of the surface

In order to study the electrostatic response of the sample uponUV/O3 exposure, we have performed ESFM on the P3OT thin films.Fig. 2 shows a set of electrostatic images before and after UVirradiation. These images were acquired simultaneously with thecorresponding topographic images shown in Fig. 1. For the pristinesample (Fig. 2A), the electrostatic image is almost flat and only abright feature is seen in the lower part of the image, whichcorresponds to the large round structure of Fig. 1A. For 1–3 minof UV irradiation (images not show) the electrostatic imagescontinue to be essentially flat and the round feature loosesbrightness. In fact, after 3 min of exposure this feature is indis-tinguishable from the rest of the image. However, after 5 min thefeature becomes darker (see lower part of Fig. 2B). The maximumcontrast of this feature is obtained after 7 min of UV irradiation

the corresponding topographic images shown in Fig. 1. (A) pristine sample (B) after

1.2 V.

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–13321330

(Fig. 2C) and is studied in detail in Fig. 3, where the histogramsshow the difference of electrostatic signal between the roundfeature, which is about 100 mV. After 10 min the round featurelooses contrast (Fig. 2D). Finally, for the higher exposures(15–30 min) bright features appear in the electrostatic images.These features are correlated with the large morphological varia-tions observed in the topographic images (Fig.1G and H). In order toquantify the variation of electrostatic properties as the samples areirradiated, the mean value of electrostatic signal has been calcu-lated for a region, where the electrostatic background is homo-geneous. As shown in Fig. 4 the electrostatic signal decreases from0 mV to about �200 mV during the first 15 min and remainsconstant thereafter. These values, even if relative measurementsbetween different parts of the sample and not absolute valuescalibrated against a fixed external voltage reference, are consistentwith the values reported by other groups for the energy distribu-tion of traps using very different experimental techniques: the peak

Fig. 3. Histogram of the electrostatic signal for the areas marked in the inset. The

inset is a zoom of Fig. 2 (C).

Fig. 4. Electrostatic signal obtained from the electrostatic images of Fig. 2. The

electrostatic value has been obtained from a zoom of 4�4 mm2 in an image region

with a homogeneous background.

displacement that we are observing (Fig. 3, 106 meV) has almostthe same value as the center of the distribution of the dominanttraps with respect to the equilibrium level reported for P3HT whichare strongly related to the exposure to oxygen (105 meV, aspreviously reported [29,30]).

In addition to nanoscale morphological and electrostatic prop-erties, also the (macroscopic) optical properties of the film weremonitored. We observe that while the pristine films have a redcolor, this color bleaches as soon as the sample is irradiated andafter 10–15 min the films becomes completely transparent. Thiseffect of photobleaching has been previously reported on MEH-PPVmolecules and indicates that the initial photooxidation products ofconjugated polymers are charged species, rather than covalentoxidation products, which suggest a strong relationship amongphotobleaching, charge separation, and persistent photoconduc-tivity of organic conjugated materials [40]. A detailed study of thesevariations of optical properties will be presented elsewhere [39].Nevertheless, we noted that the bleaching of the film starts with thefirst doses of irradiation, when neither the film morphology, nor thefilm thickness have varied significantly. Significant nanoscalemorphological and electrostatic degradation is not observed untilthe sample has been irradiated for 10–15 min, a time span whichhas been sufficient for essential complete optical degradation.

3.3. Electronic transport measurements

After each cycle of UV irradiation, an I–V measurement wasperformed on the sample as described in the experimental section.The results are summarized in Fig. 5, where an I–V plot is shown forincreasing UV irradiation cycles ranging from 1 to 15 min ofexposure. The result for the pristine sample is also plotted. TheOhmic behavior of the contacts is evident from the linearity of theplots in the applied voltage interval spanning from �1 to 1 V,which allows for a resistance calculation from the linear fits that aresuperimposed in Fig. 5. The resistance value is then converted intoconductivity using the geometrical factor correction, since the barshape and the thickness of the sample are well known. We haveused the thickness values provided by the SFM measurementswhich are showed in Table 1. The conductivity of the sample isaround 10�6 S/cm, which is of the order of magnitude previouslyreported for P3OT. A clear trend of conductivity reduction isobserved for the first seven minutes of irradiation, the linearreduction in the plot at logarithmic scale of conductivity pointsto a volume effect, where the conductivity is clearly 3D and thereduction of conjugated domains in the polymer chain by the effectof UV irradiation has a straightforward effect in conductivityreduction. This reduction trend in conductivity is coincident withthe optical decoloration of the sample as indicated in the abovesection. If we consider that the number of carriers is held constantduring this first stage of degradation, then this reduction inconductivity is owed to a reduction in mobility (s¼ne2t/m*¼nem,where m¼et/m* is the mobility in cm2/(V s) and m* the effectivemass of the carriers) and the relative degradation of transportproperties can be shown as a relative decrease in mobility (Fig. 5,right) which can be fitted to deliver an exponential decrease rate,m¼m0 exp(at), with a¼0.1670.02 s�1. The trend is broken by asudden improvement of the conductivity (two orders of magni-tude) upon irradiation of 10 min, this effect is not driven by ageometrical effect arising from thickness reduction in the calcula-tion from resistance to conductivity because it can also be clearlyseen in the I–V graph, before any correction is performed: after10 min of irradiance, the conductance of the sample is improved.We explain this increment in conductivity by oxygen doping of thesample: in the second stage of degradation, the macromolecularmodification of the polymer allows the incorporation of oxygen

Fig. 5. Left: I–V plot of the electronic transport measurements of thin films of P3OT spin cast on a glass substrate. The linear fits allow us to obtain a resistance for the sample

after each UV exposure and the conductivity obtained using the values of thickness obtained by SFM (Table 1) is presented in the inset. Right: The relative decrease of mobility

as a result of UV irradiation time.

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–1332 1331

impurities which acts as a p-type dopant for the sample, thusincreasing the carrier density, this hypothesis is further supportedby the surface charge modification observed in the electrostaticforce microscopy experiment (Fig. 3). A similar effect has beenreported recently for P3HT:PCBM blends [29]. The reduction trendin the mobility is kept as in the first stage, which explains thesubsequent reduction observed after the step for the final 15 minirradiance.

The charge transfer complex (CTC) formed by the interaction ofthe oxygen with the polymer is largely responsible for the gen-eration of charge carriers [23]. This effect is expected in allp-conjugated polymers and its effect is to increase the conductivityof the polymer, in particular, those having an amorphous nature ora very small crystallinity percentage in the bulk, which is the casefor P3OT, facilitate diffusion of oxygen into the bulk. ThereforeP3OT will be particularly susceptible to charge transfer complexformation in the presence of oxygen. This CTC is in principlereversible, but since our samples have been also exposed to UV lightduring oxygen exposure, the mobility of the polymer is degraded byphotooxidation and the change is no longer reversible. The mobilitydecreases continuously and it is never recovered. This indicatesthat the UV light should be avoided during any fabrication step of adevice, since its effect is more negative on device performance thanthe exposure to oxygen (in dark conditions).

For samples which suffered higher exposures to UV irradiation,the extremely low thickness of the sample did not allow us to getreliable ohmic contacts to the polymeric Hall bar. Also, for thesmaller thicknesses of the sample, a dimensional effect could alsobe occurring for thickness below 60 nm, which is of the order ofmagnitude of the mean free path of the charge carriers andtherefore a 2D localization effect, where the electronic transportbecomes bi-dimensional, could be playing an important role. Theeffect has been confirmed for several samples. The best way tostudy this dimensionality effect would be to perform temperaturestudies, where the temperature of the sample can be varied in acontrolled way from room temperature to cryogenic temperatures(at least 4.2 K using a liquid helium cryostat). In the presentdiscussion we have assumed that the effective mass (m*) of thecarriers in the polymer remains the same, but a decoupling of themobility and diffusion time of the carriers could also play a role ifthe effective mass varies as a consequence of the macromolecularmodification of the polymer upon degradation caused by the UVirradiation. The classical test for probing the modification of theeffective mass is to perform an alternate current (AC) excitation ofthe sample and measuring the impedance peak which marks afrequency at which the resonance with the cyclotron frequency ofthe charge carriers is produced for a particular magnetic field(wc¼eB/m*), which is applied to the sample, again this would be anice magnetotransport experiment for low temperatures.

4. Conclusion

The UV/ozone degradation on P3OT thin films was investigatedby means of non-contact-dynamic surface force microscopy andI–V transport measurements. Our experimental setup allowed us toperform a nanoscale study of the same area of the sample andtherefore we can attribute the observed changes to the effect of UVradiation and ozone. Clearly, a two stage process is produced underirradiation. We find that for the first 10–15 min of exposure themorphology and the thickness of the films do not change sig-nificantly, but color of the sample is lost, while the electrostaticproperties and the conductivity changes markedly. A decrease inmobility is observed at an exponential rate of a¼0.1670.02 s�1.After 10–15 min the film enters a second stage of degradation,there are marked changes in the morphology as well as in the filmthickness and a strong contamination by oxygen occurs producinga doping effect which increases the carrier density and thereforethe conductivity in a single step, although the mobility continuesdeteriorating.

Acknowledgements

The work was supported by the Comunidad Autonoma de laRegion de Murcia under the project ‘‘Organic Solar Cells: Frommolecular and nanoscale structure to macroscale operationaldevices’’ as well as through the project ‘‘Propiedades Nanometricasde Celulas Solares Organicas’’ (MAT2010-21267-C02) from theMICINN. A.U. acknowledges financial support from MICINNthrough project HOPE CSD2007-00007 (Consolider- Ingenio2010) and from Fundacion Seneca (CARM) grant 14898/EE10/10.

References

[1] C.J. Brabec, N.S. Sariciftci, J.C. Hummelen, Plastic solar cells, Adv. Funct. Mater.11 (2001) 15–26.

[2] T.J. Skotheim (Ed.), Handbook of Conducting Polymers, Marcel Dekker, NewYork, 1986 vol. 1 and 2.

[3] N. Camaioni, M. Catellani, S. Luzzati, A. Magliori, Morphological characterization ofpoly(3-octylthiophene) plasticizer: C60 blends, Thin Solid Films 403–404 (2002)489–494.

[4] T.J. Prosa, M.J. Winokur, J. Moulton, P. Smith, A.J. Heeger, X-ray-diffractionstudies of the three-dimensional structure within iodine-intercalated poly(3-octylthiophene), Phys. Rev. B 51 (1995) 159–168.

[5] B.S. Ong, Y. Wu, P. Liu, S. Gardner, Structurally ordered polythiophenenanoparticles for high-performance organic thin-film transistors, Adv. Mater.17 (2005) 1141–1144.

[6] T. Kaniowski, S. Niziol, J. Sanetra, M. Trznadel, A. Pron, Optical studies ofregioregular poly(3-octylthiophene)s under pressure, Synth. Met. 94 (1998)111–114.

[7] R. Singh, J. Kumar, R.K. Singh, A. Kaur, K.N. Sood, R.C. Rastogi, Effect of thermalannealing on surface morphology and physical properties of poly(3-octylthio-phene) films, Polymer 46 (2005) 9126–9132.

J. Abad et al. / Solar Energy Materials & Solar Cells 95 (2011) 1326–13321332

[8] M. Manceau, A. Rivaton, J.-L. Gardette, S. Guillerez, N. Lemaıtre, The mechanismof photo- and thermooxidation of poly(3-hexylthiophene) (P3HT) reconsid-ered, Polym. Degrad. Stab. 94 (2010) 898–907.

[9] F.C. Krebs, K. Norrman, Analysis of the failure mechanism for a stable organicphotovoltaic during 10,000 h of testing, Prog. Photovolt: Res. Appl. 15 (2007)697–712.

[10] M. Jørgensen, K. Norrman, F.C. Krebs, Stability/degradation of polymer solarcells, Sol. Energy Mater. Sol. Cells 92 (2008) 686–714.

[11] S. Chambon, A. Rivaton, J. Gardette, M. Firon, Photo- and thermal degradation ofMDMO–PPV:PCBM blends, Sol. Energy Mater. Sol. Cells 91 (2007) 394–398.

[12] M. Manceau, J. Gaume, A. Rivaton, J.-L. Gardette, G. Monier, L. Bideux, Furtherinsights into the photodegradation of poly(3-hexylthiophene) by means ofX-ray photoelectron spectroscopy, Thin Solid Films 518 (2010) 7113–7118.

[13] M. Manceau, S. Chambon, A. Rivaton, J.-L. Gardette, S. Guillerez, N. Lemaıtre,Effects of long-term UV-visible light irradiation in the absence of oxygen on P3HTand P3HT:PCBM blend, Sol. Energy Mater. Sol. Cells 94 (2010) 1572–1577.

[14] E. Kymakis, I. Alexandrou, G.A.J. Amaratunga, High open-circuit voltagephotovoltaic devices from carbon-nanotube-polymer composites, J. Appl.Phys. 93 (2003) 1764–1768.

[15] S. Bhattacharyya, E. Kymakis, G.A.J. Amaratunga, Photovoltaic properties of dyefunctionalized single-wall carbon nanotube/conjugated polymer devices,Chem. Mater. 16 (2004) 4819–4823.

[16] J.F. Rabek, Polymer Photodegradation: Mechanism and Experimental Methods,Chapman & Hall, 1995.

[17] S. Hamid S, Handbook of Polymer Degradation, 2nd ed., Marcel Dekker, Inc, 2000.[18] M. Chabinyc, R.A. Street, J.E. Northrup, Effects of molecular oxygen and ozone

on polythiophene-based thin-film transistors, Appl. Phys. Lett. 90 (2007)123508–123511.

[19] J. Nowaczyk, P. Olszowy, P. Cysewski, A. Nowaczyk, W. Czerwinski, Ozoniza-tion of electronic conducting polymers, part III: the action of ozone on poly[3-pentylthiophene] film, Polymer Degrad. Stab. 93 (2008) 1275–1283.

[20] M.T. Loponen, T. Taka, J. Laakso, K. Vakiparta, K. Suuronen, P. Valkeinen,J.-E. Osterholm, Doping and dedoping processes in poly(3-alkylthiopehenes),Synth Metals 41 (1991) 479–484.

[21] T. Taka, M.T. Loponen, J. Laakso, K. Suuronen, P. Valkeinen, J.E. Osterholm,Degradation studies of doped and dedoped poly(3-octyl thiophene), SyntheticMetals 41 (1–2) (1991) 567–570.

[22] M.S.A. Abdou, F.P. Orfino, Z.W. Xie, M.J. Deen, S. Holdcroft, Reversible chargetransfer complex between molecular oxygen and poly(3-alkylthiophenes),Adv. Mater. 6 (1994) 838–841.

[23] M.S.A. Abdou, F.P. Orfino, Y. Son, S. Holdcroft, Interaction of oxygen withconjugated polymers: charge transfer complex formation with poly(3-alkylthiophenes), J. Am. Chem. Soc. 119 (1997) 4518–4524.

[24] Y. Rodriguez-Lazcano, H. Martinez, Emission spectroscopic analysis of poly(3-octylthiophene) exposed to plasmas, J. Appl. Pol. Sci. 105 (2007) 2947–2954.

[25] H.H. Liao, C.M. Yang, C.C. Liu, S.F. Horng, H.F. Meng, J.T. Shy, Dynamics andreversibility of oxygen doping and de-doping for conjugated polymer, J. Appl.Phys. 103 (2008) 104506–104513.

[26] E.J. Meijer, C. Detcheverry, P.J. Baesjou, E. van Veenendaal, D.M. de Leeuw,T.M. Klapwijk, Dopant density determination in disordered organic field-effecttransistors, J. Appl. Phys. 93 (2003) 4831–4835.

[27] K. Norrman. Stability and Degradation of Polymer Solar cells. In: Proceedings ofISOS-3, 21–29 October, 2010, Risø (DK), (Invited conference contribution).

[28] Liadaurea M. Rogerio Valaski, Liliana Moreira, Micaroni, A.Hummelgen Ivo, Theelectronic behavior of poly(3-octylthiophene) electrochemically synthesizedonto Au substrate, Brazilian J. Phys. 33 (2003) 392–397.

[29] J. Schafferhans, A. Baumann, A. Wagenpfahl, C. Deibel, V. Dyakonov, Oxygendoping of P3HT:PCBM blends: influence on trap states, charge carrier mobilityand solar cell performance, Org. Electron. 11 (2010) 1693–1700.

[30] J. Schafferhans, A. Baumann, C. Deibel, V. Dyakonov, Trap distribution and theimpact of oxygen-induced traps on the charge transport in poly 3-hexylthio-phene, Appl. Phys. Lett. 93 (2008) 093303–093305.

[31] C.K. Lu, H.F. Meng, Hole doping by molecular oxygen in organic semiconduc-tors: band-structure calculations, Phys. Rev. B 75 (2007) 235206–235211.

[32] C.K. Lu, S.T. Pi, H.F. Meng, Effect of defect-enhanced molecular oxygenadsorption on the imbalance of hole versus electron mobility in conjugatedpolymers, Phys. Rev. B 75 (2007) 195206–195210.

[33] M.R. Lilliedal, A.J. Medford, M.V. Madsen, K. Norrman, F.C. Krebs., The effect ofpost-processing treatments on inflection points in current–voltage curves ofroll-to-roll processed polymer photovoltaics, Sol. Energy Mater. Sol. Cells 94(2010) 2018–2031.

[34] J. Mardalen, H.J. Fell, E.J. Samuelsen, E. Bakken, P.H.J. Carlsen, M.R. Andersson,X-ray structural studies of various octyl-substituted polythiophenes macro-mol, Chem. Phys. 196 (1995) 553–565.

[35] I. Horcas, R. Fernandez, J.M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero,A.M. Baro, WSXM: a software for scanning probe microscopy and a tool fornanotechnology, Rev. Sci. Inst. 78 (2007) 013705–013713.

[36] C.H. Lei, A. Das, M. Elliott, J.E. Macdonald, Quantitative electrostatic forcemicroscopy-phase measurements, Nanotechnology 15 (2004) 627–634.

[37] E. Palacios-Lidon, B. Perez-Garcia, J. Colchero, Enhancing dynamic scanningforce microscopy in air: as close as possible, Nanotechnology 20 (2009)085707–085711.

[38] B. Perez-Garcia, J. Abad, A. Urbina, J. Colchero, E. Palacios-Lidon, Surfacepotential domains on lamellar P3OT structures, Nanotechnology 19 (2008)065709–065714.

[39] J. Abad, A. Urbina, J. Colchero, Organic Electronics Journal, submitted forpublication.

[40] S.-J. Park, A.J. Gesquiere, J. Yu, P.F. Barbara., Charge injection and photooxida-tion of single conjugated polymer molecules, J. Am. Chem. Soc. 126 (2004)4116–4117.