high-efficiency inverted dithienogermole–thienopyrrolodione-based polymer solar cells

High-efficiency inverted dithienogermole–thienopyrrolodione-based polymer solar cellsCephas E. Small1†, Song Chen1†, Jegadesan Subbiah1, Chad M. Amb2, Sai-Wing Tsang1, Tzung-Han Lai1,

John R. Reynolds2* and Franky So1*

Inverted polymer bulk heterojunction solar cells have received a great deal of attention because of their compatibility withlarge-scale roll-to-roll processing. The inverted cell geometry has the following structure: substrate (rigid orflexible)/indium tin oxide/electron-transporting layer/photoactive layer/hole-transporting layer/top anode. Solution-processed metal-oxide films, based on materials such as ZnO and TiO2, are typically used as the electron-transportinglayers. Here, we demonstrate enhanced charge collection in inverted polymer solar cells using a surface-modified ZnO–polymer nanocomposite electron-transporting layer. Using this approach, we demonstrate inverted polymer solar cellsbased on a low-bandgap polymer with an alternating dithienogermole–thienopyrrolodione repeat unit (PDTG–TPD) withcertified power conversion efficiencies of 7.4%. To our knowledge, this is the highest efficiency reported to date forpolymer solar cells with a device architecture compatible with the roll-to-roll process.

Extensive efforts have been directed at developing polymer bulkheterojunction (BHJ) solar cells because of their potential forlow-cost energy harvesting1,2. High-efficiency, laboratory-scale

polymer solar cells based on low-bandgap polymers have beenreported with Air Mass 1.5 Global (AM 1.5G) power conversionefficiencies (PCEs) of 4.7–6.8% (refs 3–7). The device geometry oftypical laboratory-scale polymer solar cells comprises a bottomindium tin oxide (ITO) anode, an anode interfacial layer, a photo-active layer and a low-work-function top metal cathode. Becausevacuum deposition of low-work-function metals is required forthese top cathode devices, it is not viable to use this device architec-ture in large-scale roll-to-roll (R2R) processing.

To avoid the low-work-function metals used in such devices, twostrategies have been adopted. In the first, the low-work-functionmetal is replaced with a transition-metal oxide such as TiOx (ref. 8)or ZnO (ref. 9). In the second, the polarity of the BHJ devices10–13 isinverted. Specifically, these so-called ‘inverted cells’ have an oxide elec-tron-transporting layer (ETL)-coated ITO as the bottom cathode and ascreen-printed silver layer as the top anode; to date, this device archi-tecture is prototypical for R2R processing10–13. To fabricate the ETL,solution-processed metal oxides have been widely used in large-scaleinverted solar cells to reduce the ITO work function. In particular,ZnO colloidal nanoparticles are used for the ETL because of theirlow work function, high electron mobility and optical transparency,as well as their ease of synthesis2,10–15. However, the major challengesin using ZnO nanoparticle films as ETLs are the presence of defectswith adsorbed oxygen16,17 and the poor spatial distribution of thenanoparticles over a large area18–20. Accordingly, there is a need todevelop low-defect and uniform ZnO films so as to realize high-effi-ciency inverted polymer solar cells.

Here, we report a new method for enhancing charge collection ininverted polymer BHJ solar cells using a ZnO–poly(vinyl pyrroli-done) (PVP) composite sol–gel film as the ETL, and demonstrateinverted polymer solar cells that operate with laboratory-measuredPCEs in excess of 8% and certified efficiencies of 7.4% under AM1.5G illumination at 100 mW cm22. The composite film, termed

‘ZnO–PVP nanocomposite’ for clarity, consists of ZnO nanoclusterswhose growth is mediated by a PVP polymeric matrix21. ZnO–PVPnanocomposite films have been studied previously in relation tochemical/biosensing applications22–25 and have the followingadvantages over conventional ZnO sol–gel films. First, the ZnOnanocluster size and its concentration can be tuned by controllingthe Zn2þ/PVP ratio22,23. Second, the distribution of ZnO nanoclus-ters in the PVP polymer is uniform compared to the aggregationobserved in ZnO sol–gel films without PVP22,23. We hypothesizedthat inverted solar cells using this composite would demonstrateenhanced device performance. Furthermore, because the sol–gelprocessing for the ZnO–PVP nanocomposite is performed in air,this approach to depositing the ZnO ETLs is compatible withlarge-scale R2R processes.

We recently reported the synthesis and BHJ solar cell performanceof a dithienogermole (DTG) containing an alternating conjugateddonor–acceptor polymer, in which N-octylthienopyrrolodione (TPD)was used as the acceptor26. Inverted polydithienogermole-thienopyrro-lodione (PDTG-TPD):[6,6]-phenyl-C71-butyric acid methyl ester(PC71BM) solar cells demonstrated a higher short-circuit currentdensity (Jsc) and fill factor (FF) than devices with an analogous poly-dithienosilole-containing polymer (PDTS–TPD), leading to invertedpolymer solar cells with PCEs of 7.3% (ref. 27). To determinewhether our ZnO–PVP nanocomposite films would enhance chargecollection for inverted solar cells, we fabricated inverted PDTG–TPD:PC71BM BHJ solar cells using the nanocomposite as an ETL.The following inverted device geometry was used for our study:ITO/ZnO–PVP nanocomposite/PDTG–TPD:PC71BM/anode inter-facial layer/silver. For the anode interfacial layer, molybdenum trioxide(MoO3) was used, which has a better charge extraction efficiency thanthe more commonly used poly(ethylenedioxythiophene) doped withpoly(styrene-sulphonate) (PEDOT:PSS)26,28–31. Although both MoO3and silver are thermally evaporated in our devices, these electrodematerials are compatible with R2R processing because MoO3 can becoated from a nanoparticle suspension32 and silver paste can bescreen-printed10–13.

1Department of Materials Science and Engineering, University of Florida, Box 117200, Gainesville, Florida 32611, USA, 2The George and Josephine ButlerPolymer Research Laboratory, Department of Chemistry, Center for Macromolecular Science and Engineering, University of Florida, Box 117200, Gainesville,Florida 32611, USA; †These authors contributed equally to this work. *e-mail: [email protected]; [email protected]

ARTICLESPUBLISHED ONLINE: 18 DECEMBER 2011 | DOI: 10.1038/NPHOTON.2011.317

NATURE PHOTONICS | VOL 6 | FEBRUARY 2012 | www.nature.com/naturephotonics 115

© 2012 Macmillan Publishers Limited. All rights reserved.

mailto:[email protected]


http://www.nature.com/doifinder/10.1038/nphoton.2011.317

www.nature.com/naturephotonics

The photo J–V characteristics for inverted PDTG–TPD:PC71BMsolar cells were measured under AM 1.5G solar illumination at100 mW cm22. The photovoltaic (PV) performance results for theinverted cells with ZnO–PVP nanocomposites are shown inFig. 1. On initial light exposure the inverted solar cells had a lowFF of 25.5% and Jsc of 10.9 mA cm22. With continuous illumina-tion, device performance was enhanced significantly over time33.After �10 min of light soaking, an enhanced FF of 63.7% and Jscof 12.9 mA cm22 were obtained, resulting in an average PCE of7.0%. Previously, we reported inverted PDTG–TPD-basedpolymer solar cells with a FF of �68% using colloidal ZnO nanopar-ticles as the ETL26. We suspected that by using ZnO-PVP compositeas the ETL, the ZnO-PVP surface would be compositionally rich inPVP, creating a contact barrier between the ZnO nanoclusters andPC71BM leading to the lower FF of our present devices.

To ensure a good contact between the ZnO nanoclusters andPC71BM, we performed UV-ozone treatment on the ZnO–PVPnanocomposite films to remove PVP from the surface. Previouswork has shown that UV-ozone treatment can remove PVP on col-loidal nanoparticle film surfaces34. The removal of PVP did not alterthe size, shape or distribution of the nanoclusters in the films34,35.Based on these findings, we believed that the UV-ozone treatmentwould improve electronic coupling between the photo-active layerand the ZnO nanoclusters.

The photo J–V characteristics for the inverted PDTG–TPD:PC71BM solar cells with UV-ozone treated ZnO–PVP nano-composites are shown in Fig. 2a. All devices were tested underinitial light exposure, and no additional light soaking was appliedto the devices. Additional information regarding the light-soakingmechanism is presented in the Supplementary Information(Supplementary Fig. 1). The ZnO–PVP nanocomposite films wereUV-ozone treated for 5, 10, 20 and 30 min, leading to significantenhancements in the Jsc and FF values for the inverted PDTG–TPD:PC71BM solar cells compared to cells with as-prepared nano-composite films. Table 1 summarizes the device performance forinverted solar cells with treated ZnO–PVP nanocomposite films.UV-ozone treating the ZnO–PVP nanocomposite films for10 min led to an optimal device with enhancements in both Jscand FF compared to the light-soaked devices without UV-ozonetreatment, and resulting in an average PCE of 8.1%. This averagePCE of 8.1+0.4% is based on measurements from 102 fabricated

solar cells; the corresponding device performance is shown as a his-togram in the Supplementary Information (Supplementary Fig. 2).Our best device had a Jsc of 14.4 mA cm22, Voc of 0.86 V, FF of68.8% and PCE of 8.5%. For devices with ZnO–PVP nanocompositefilms that had been UV-ozone treated for less than or more than10 min, a reduction in FF was observed. For the shorter treatment,we attribute this reduction in FF to incomplete removal of the PVPfrom the surface of the composite film. For the longer treatment,excess oxygen is present on the ZnO film surface, which reducesthe electron extraction efficiency. Based on these findings, we con-clude that removal of extra PVP from the ZnO–PVP nanocompositefilm surface by UV-ozone treatment greatly enhances the chargecollection efficiency of these devices.

To confirm the accuracy of the photo J–V measurements, theexternal quantum efficiency (EQE) spectra for the solar cells withas-prepared and 10 min UV-ozone treated ZnO–PVP nanocompo-site films were measured; these are compared in Fig. 2b. Anenhanced efficiency is observed throughout the full spectral rangefrom 350–700 nm for cells with UV-ozone treated ZnO–PVP

0.0 0.2 0.4 0.6 0.8 1.0−14

−12

−10

−8

−6

−4

−2

0

2

4

6C

urre

nt d

ensi

ty (m

A c

m−2

)

Voltage (V)

Control—initial light exposure10 s light soaking5 min light soaking10 min light soaking

Figure 1 | Effect of light soaking on device performance for inverted solar

cells with an as-prepared ZnO–PVP nanocomposite ETL. Photo J–V

characteristics for inverted PDTG–TPD:PC71BM solar cells, highlighting the

effect of prolonged light soaking on device performance under AM 1.5G

solar illumination at 100 mW cm22.

0.0 0.2 0.4 0.6 0.8 1.0−16

−14

−12

−10

−8

−6

−4

−2

0

2

4

6a

b

5 min UV-ozone10 min UV-ozone20 min UV-ozone30 min UV-ozone

Cur

rent

den

sity

(mA

cm

−2)

Voltage (V)

300 400 500 600 700 800 9000

10

20

30

40

50

60

70

80

90

UV-ozone treatedZnO compositeAs-preparedZnO composite

EQE

(%)

Wavelength (nm)

Figure 2 | Enhanced device performance in inverted PDTG–TPD:PC71BM

solar cells by UV-ozone treatment of the ZnO–PVP nanocomposite ETL.

a, Photo J–V curves of inverted PDTG–TPD:PC71BM solar cells with

UV-ozone treated ZnO–PVP nanocomposite films as ETLs for various

treatment times (5, 10, 20 and 30 min) under initial AM 1.5G solar

illumination at 100 mW cm22. b, Corresponding EQE for devices with 10 min

UV-ozone treated ZnO–PVP nanocomposite films. The EQE spectrum for

devices with as-prepared composite films is shown for comparison.

ARTICLES NATURE PHOTONICS DOI: 10.1038/NPHOTON.2011.317

NATURE PHOTONICS | VOL 6 | FEBRUARY 2012 | www.nature.com/naturephotonics116




nanocomposite films when compared to cells without UV-ozonetreatment. The maximum EQE for the optimized invertedPDTG–TPD:PC71BM solar cell with UV-ozone treated nanocom-posite films was 73.6%. The Jsc value was then calculated by integrat-ing the EQE data with the AM 1.5G spectrum. The calculated Jscvalue of 14.5 mA cm22 is in good agreement with the directlymeasured Jsc value.

Encapsulated devices with UV-ozone treated ZnO–PVP nano-composite films were then sent to NEWPORT Corporation for cer-tification. The photo J–V characteristics and the corresponding solarcell parameters are shown in Fig. 3. A PCE of 7.4+0.2% was certi-fied for the devices. Although this certified efficiency is 9% less thanthat measured in our laboratory because of a reduction in Jsc and FFin the certified device, we attribute the reduction in PCE in the cer-tified cells to degradation because of a non-optimized encapsulationprocess. The devices were retested in our laboratory after certifica-tion and we obtained an efficiency (7.2%) comparable to thecertified results.

We sought to further investigate the effect of UV-ozone treat-ment on ZnO–PVP composite films and the corresponding photo-voltaic device stability. As shown in Supplementary Fig. 3, after lightsoaking, the performance of the device with untreated ZnO–PVPnanocomposite ETL reduced significantly over time, showing thatthe enhancement in Jsc and FF due to light soaking was only tem-porary33. In contrast, the device performance observed for solarcells with UV-ozone treated ZnO–PVP nanocomposite ETLs hadmuch better stability. In fact, we observed no measurable changesin Jsc , FF or PCE over a period of one month provided that theencapsulated devices were stored in a nitrogen glovebox. Thus,UV-ozone treatment of the ZnO–PVP nanocomposite films pro-vides enhanced stability in the inverted solar cells.

The enhanced device performance with UV-ozone treated ZnO–PVP nanocomposite films is believed to be attributable to themodified surface composition promoting charge collection. Thenanoscale surface morphologies of the as-prepared and UV-ozonetreated ZnO–PVP nanocomposite films were investigated byatomic force microscopy (AFM). Figure 4a,b shows the three-dimensional surface topography images for the nanocompositefilm before and after UV-ozone treatment. The ZnO–PVP nano-composite film shows an increase in r.m.s. roughness from7.07 nm to 9.18 nm following UV-ozone treatment, suggestingthat PVP is removed with this treatment, leaving the ZnO nanoclus-ters exposed at the surface. This removal of the PVP is more clearlyshown in the AFM phase images in Fig. 4c,d. For the nanocompo-site film without UV-ozone treatment, no nanoclusters can beobserved, indicating that the surface is covered by a thin layer ofPVP. However, the phase image for the UV-ozone treated ZnO–PVP nanocomposite film shows that the PVP domain size hasbeen reduced to 50–100 nm and that the ZnO nanoclusters arenow exposed on the surface. The PVP-rich and ZnO nanocluster-rich surfaces of the nanocomposite films before and after UV-ozone treatment, respectively, are shown schematically in Fig. 4e,f.

We also investigated whether the removal of PVP from the nano-composite film surface altered the film thickness by performingstep-height measurements for the films before and afterUV-ozone treatment. The average thickness of the nanocomposite

film shows negligible change after 10 min UV-ozone treatment,which provides further evidence that PVP was removed only atthe surface of the ZnO–PVP nanocomposite film upon UV-ozonetreatment. Furthermore, to illustrate the surface morphology ofthe ZnO–PVP nanocomposite films before and after UV-ozonetreatment, 5-mm-scale phase images are shown in SupplementaryFig. 4. These images show that the removal of PVP by UV-ozonetreatment significantly alters the surface morphology of the nano-composite film. This change in surface morphology from beingPVP-rich before UV-ozone treatment to ZnO-rich after treatmentsupports our idea that the removal of PVP from the nanocompositefilm by UV-ozone treatment leads to improved charge collection inour inverted polymer solar cells due to better electronic couplingbetween the ZnO nanoclusters within the nanocomposite filmand PC71BM in the active layer.

To further confirm that the compositional changes determinedfrom the AFM data were due to the removal of PVP, X-ray photo-emission spectroscopy (XPS) was performed on the ZnO–PVPnanocomposite films. Considering the period of UV-ozone treat-ment required to optimize device performance, it is believed thatsome changes in the chemical composition of the ZnO might beplausible. The core-level XPS spectra for C 1s, O 1s and Zn 2pwere measured for the as-prepared and 10 min UV-ozone treatedZnO–PVP nanocomposite films. The binding energies were cali-brated by taking the C 1s peak (284.6 eV) as a reference. The O 1sXPS spectra for as-prepared and UV-ozone treated ZnO–PVPnanocomposite films are shown in Fig. 5a. UV-ozone treatmentincreased the relative magnitude of the peak at 531.4 eV (corre-sponding to the oxygen atoms bonded to the zinc in the ZnOmatrix22,23,36) by �37%. Thus the number of Zn–O bonds in thewurtzite structure of ZnO at the surface of the film is increased.UV-ozone treatment also increased the relative magnitude ofthe peak at �530.0 eV, which corresponds to O22 ions present in

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.90.0000

0.0001

0.0002

0.0003

0.0004

Cur

rent

(A)

Voltage (V)

Voc = 0.8652 V Vmax = 0.7143 V

Isc = 0.3950 mA Imax = 0.31 mA

Jsc = 12.993 mA cm−2 Pmax = 0.224 mW

Fill factor = 65.7% Efficiency = 7.4%

Figure 3 | Certified I–V characteristics for an inverted PDTG–TPD:PC71BM

solar cell with 10 min UV-ozone treated ZnO–PVP nanocomposite ETL. The

device was certified by NEWPORT Corporation. A light mask with an area of

0.0304 cm2 was used to define the device area.

Table 1 | Averaged solar cell performance for inverted PDTG–TPD:PC71BM devices with 5, 10, 20 or 30 min UV-ozone treatedZnO–PVP composite ETLs under initial AM 1.5G solar illumination.

UV-ozone treatment time Jsc (mA cm22) Voc (V) FF (%) Average PCE Best PCE

5 min 13.9+0.1 0.85+0.005 56.0+3.8 6.6+0.5 7.110 min 14.0+0.4 0.86+0.003 67.3+1.5 8.1+0.4 8.520 min 14.1+0.2 0.86+0.003 64.8+0.8 7.8+0.2 8.030 min 14.0+0.1 0.86+0.003 61.9+1.6 7.5+0.3 7.6

NATURE PHOTONICS DOI: 10.1038/NPHOTON.2011.317 ARTICLES





the porous ZnO clusters, but not chemically bonded to zinc in theZnO wurtzite structure22,23. Figure 5b shows the Zn 2p3/2 XPSspectra for the as-prepared and UV-ozone treated ZnO–PVP nano-composite films. The intensity of the peak at 1,021.6 eV, whichcorresponds to the Zn–O bonds22,23,36,37, increases after UV-ozonetreatment. These results are in agreement with the result for theO 1s XPS spectra. Based on the O 1s and Zn 2p XPS spectra, we con-clude that the chemical composition of the ZnO nanoclusters on thesurface of the nanocomposite film has become oxygen-rich afterUV-ozone treatment.

The atomic concentrations of carbon, oxygen and zinc in the as-prepared and 10 min UV-ozone treated ZnO–PVP nanocompositefilms based on the C 1s, O 1s and Zn 2p XPS spectra are summarizedin Fig. 5c. The atomic concentration of carbon from the PVP in thenanocomposite is significantly reduced by UV-ozone treatment(from 38.2% to 15.7%). Conversely, the atomic concentrations ofoxygen and zinc present in the nanocomposite film both increasefrom 28.5% and 33.3% for the untreated film to 31.6% and 52.6%,respectively, for the treated film. The relatively smaller increasein oxygen atomic concentration compared to zinc is due to the com-petition between the increases in oxygen content arising fromUV-ozone treatment and the decrease in oxygen content comingfrom the removal of PVP. These results strongly support our asser-tion that UV-ozone treatment removes PVP from the surface of theZnO–PVP nanocomposite film.

Finally, we studied the effect of UV-ozone treatment on opticaltransmission for the as-prepared and treated ZnO–PVP nanocom-posite films. UV-vis-NIR transmission spectra for these films areshown in Supplementary Fig. 5. Following UV-ozone treatment, a6–10% increase in transmission across the entire visible spectrumwas observed in the nanocomposite film. We attribute this increaseto the changes in the effective index of refraction of the nanocom-posite film upon UV-ozone treatment. We note that the increasein optical transparency is less than the enhancement observed inJsc for inverted solar cells with UV-ozone treated ZnO–PVP nano-composite films. Therefore, although the increase in optical trans-parency contributes to the enhancement in Jsc , the improvedcharge collection due to enhanced electronic coupling betweenZnO nanoclusters in the nanocomposite film and PC71BM in theactive layer is primarily responsible for this enhancement.

To conclude, improved charge collection efficiency has beendemonstrated in inverted PDTG–TPD BHJ solar cells using aUV-ozone treated ZnO–PVP nanocomposite film as the ETL toobtain organic polymer solar cells with PCEs in excess of 8%under AM 1.5G illumination at 100 mW cm22. The use of PVPas an organic capping molecule and polymeric matrix for ZnO pro-duced electron-transporting nanocomposite films, which had excel-lent film-forming characteristics. We found that UV-ozonetreatment was required to remove PVP from the surface of thefilm, and consequently exposed the ZnO nanoclusters to the film

1

2

3

4

(μm)

1

2

1.00

0.75

0.50

0.25

0

1.00

a

c

e f

b

d

0.75

0.50

0.25

01.00

ZnO nanoclusters

PVP chains

0.750.500.2501.000.750.500.250

3

4

50.0

25.0

0.0

10.0

5.0

0.0

(μm)

(μm) (μm)

(nm)

(deg)

Figure 4 | Tapping-mode AFM images for ZnO–PVP nanocomposite films used in device fabrication (optimized conditions). Appropriate scales are shown

on the right. a,b, Three-dimensional topography images for the as-prepared and 10 min UV-ozone treated ZnO–PVP nanocomposite films show the change in

surface roughness upon UV-ozone treatment. c, Phase image for the as-prepared nanocomposite film, showing a PVP-rich surface on which no ZnO

nanoclusters can be seen. d, Phase image for the UV-ozone treated nanocomposite film, showing a ZnO nanocluster-rich surface, with more ZnO

nanoclusters exposed on the surface. e,f, Schematic images of the as-prepared and UV-ozone treated ZnO–PVP nanocomposite films, emphasizing the

PVP-rich and ZnO-rich surfaces before and after UV-ozone treatment.






surface. Using our UV-ozone treated ZnO–PVP nanocompositeETL, inverted PDTG–TPD solar cells with laboratory-measuredPCE values higher than 8% have been realized due to improved

charge collection by the nanocomposite film. To our knowledge,this PCE is the highest yet reported for inverted polymer solarcells that have potential for large-scale manufacturing.

MethodsThe detailed synthesis, polymer characterization and device fabrication and testingfor PDTG–TPD:PC71BM and PDTS–TPD:PC71BM have been reported elsewhere4.The PC71BM used for solar cell fabrication was purchased from Solenne. Polymersand PC71BM were dissolved in chlorobenzene with a 1:1.5 (8 mg ml21:12 mg ml21)weight ratio, and 5% volume ratio of 1,8-diiodooctane (DIO) was added as aprocessing additive before use. The ZnO–PVP nanocomposite was prepared from aprecursor, in which zinc acetate dihydrate (Zn(CH3COO)2

.2H2O, Aldrich, 99.9%,11 mg ml21) and PVP (2.5 mg ml21) were dissolved in ethanol. Ethanolamine wasadded to the precursor as a stabilizer in equal molar concentration to zinc acetatedihydrate. The ZnO–PVP precursor was spin-coated on ITO-coated glass substrates,which were first cleaned with detergent, ultrasonicated in deionized water, acetoneand isopropyl alcohol, and subsequently dried under nitrogen. The films wereannealed at 200 8C for 40 min in air. After annealing and slow-cooling to roomtemperature, the ZnO–PVP composite films were UV-ozone treated using aUV-ozone cleaner. The film thicknesses for the ZnO–PVP composite film beforeand after UV-ozone treatment were 36 nm and 33 nm, respectively. The polymer–fullerene solutions were then spin-coated and the resulting film with a thickness of105 nm was annealed at 80 8C for 30 min. Finally, thin films of MoO3 (10 nm) andsilver (100 nm) were deposited through shadow masks by thermal evaporation. Theactive area of the device was 4.6 mm2. For PV measurements, a light mask with anarea of 3.04 mm2 was used to define the active area of the device. Devicecharacterization was carried out in air after encapsulation using an AM 1.5G solarsimulator with an irradiation intensity of 100 mW cm22. The beam spot used forEQE measurements was smaller than the device area to ensure accuracy ofthe measurement.

Received 25 September 2011; accepted 3 November 2011;published online 18 December 2011

References1. Li, G., Chu, C. W., Shrotriya, V., Huang, J. & Yang, Y. Efficient inverted polymer

solar cells. Appl. Phys. Lett. 88, 253503 (2006).2. Chen, L. M., Hong, Z. R., Li, G. & Yang, Y. Recent progress in polymer solar cells:

manipulation of polymer:fullerene morphology and the formation of efficientinverted polymer solar cells. Adv. Mater. 21, 1434–1449 (2009).

3. Peet, J. et al. Efficiency enhancement in low-bandgap polymer solar cells byprocessing with alkane dithiols. Nature Mater. 6, 497–500 (2007).

4. Chen, H-Y. et al. Polymer solar cells with enhanced open-circuit voltage andefficiency. Nature Photon. 3, 649–653 (2009).

5. Hou, J. et al. Synthesis of a low band gap polymer and its application in highlyefficient polymer solar cells. J. Am. Chem. Soc. 131, 15586–15587 (2009).

6. Hou, J., Chen, H-Y., Zhang, S., Li, G. & Yang, Y. Synthesis, characterization, andphotovoltaic properties of a low band gap polymer based on a silole-containingpolythiophenes and 2,1,3-benzothiadiazole. J. Am. Chem. Soc. 130,16144–16145 (2008).

7. Li, Z. et al. Alternating copolymers of cyclopenta[2,1-b;3,4-b′]diothiophene andthieno[3,4-c]pyrrole-4,6-dione for high performance polymer solar cells. Adv.Funct. Mater. 21, 3331–3336 (2011).

8. Park, S. H. et al. Bulk heterojunction solar cells with internal quantum efficiencyapproaching 100%. Nature Photon. 26, 297–303 (2009).

9. Yip, H-L., Hau, S. K., Baek, N. S., Ma, H. & Jen, A. K-Y. Polymer solar cells thatuse self-assembled-monolayer-modified ZnO/metals as cathodes. Adv. Mater.20, 2376–2382 (2008).

10. Krebs, F. C. All solution roll-to-roll processed polymer solar cells free fromindium-tin-oxide and vacuum coating steps. Org. Electron. 10, 761–768 (2009).

11. Krebs, F. C. Polymer solar cell modules prepared using roll-to-roll methods:knife-over-edge coating, slot-die coating and screen printing. Sol. Energy Mater.Sol. Cells 93, 465–475 (2009).

12. Krebs, F. C., Gevorgyan, S. A. & Alstrup, J. A roll-to-roll process to flexiblepolymer solar cells: model studies, manufacture and operational stability studies.J. Mater. Chem. 19, 5442–5451 (2009).

13. Krebs, F. C., Tromholt, T. & Jorgenson, M. Upscaling of polymer solar cellfabrication using full roll-to-roll processing. Nanoscale 2, 873–886 (2010).

14. Beek, W. J. E. et al. Hybrid zinc oxide polymer bulk heterojunction solar cells.J. Phys. Chem. B 109, 9505–9516 (2005).

15. Hau, S. K. et al. Air-stable inverted flexible polymer solar cells using zinc oxidenanoparticles as an electron selective layer. Appl. Phys. Lett. 92, 253301 (2008).

16. Guo, L. et al. Highly monodisperse polymer-capped ZnO nanoparticles:preparation and optical properties. Appl. Phys. Lett. 76, 2901–2903 (2000).

17. Mahamuni, S. et al. ZnO nanoparticles embedded in polymeric matrices.Nanostruct. Mater. 7, 659–666 (1996).

18. Cavaleri, J. J. et al. Femtosecond study of the size-dependent charge carrierdynamics in ZnO nanocluster solutions. Chem. Phys. 103, 5378–5386 (1995).

535 534 533 532 531 530 529 528 527

As-prepared ZnO compositeUV-ozone treated ZnO composite

Inte

nsity

(a.u

.)

Binding energy (eV)

O 1s

a

b

c

1,026 1,024 1,022 1,020 1,018 1,016


Inte

nsity

(a.u

.)

Binding energy (eV)

Zn 2p

C 1s O 1s Zn 2p3/20

5

10

15

20

25

30

35

40

45

50

55

60

Ato

mic

con

cent

ratio

n (%

)


Figure 5 | XPS data for the as-prepared and 10 min UV-ozone treated

ZnO–PVP nanocomposite films. a,b, O 1s (a) and Zn 2p (b) XPS spectra for

the as-prepared and 10 min UV-ozone treated ZnO–PVP nanocomposite

films. c, Atomic concentrations of carbon, zinc and oxygen before and after

UV-ozone treatment based on the corresponding XPS spectra.

NATURE PHOTONICS DOI: 10.1038/NPHOTON.2011.317 ARTICLES





19. Sakohara, S. et al. Visible luminescence and surface properties of nanosized ZnOcolliods prepared by hydrolyzing zinc acetate. J. Phys. Chem. B 102,10169–10175 (1998).

20. Monticone, S. et al. Complex nature of the UV and visible fluorescence ofcolloidal ZnO nanoparticles. J. Phys. Chem. B 102, 2854–2862 (1998).

21. Guo, L. et al. Synthesis and characterization of poly(vinylpyrrolidone)-modifiedzinc oxide nanoparticles. Chem. Mater. 12, 2268–2274 (2000).

22. Du, T. B. & Ilegbusi, O. J. Synthesis and morphological characterization onPVP/ZnO nano hybrid films. J. Mater. Sci. Lett. 39, 6105–6109 (2004).

23. Du, T. B., Song, H. W. & Ilegbusi, O. J. Sol-gel derived ZnO/PVP nanocompositethin film for superoxide radical sensor. Mater. Sci. Eng. C 27, 414–420 (2007).

24. Singla, M. L. et al. Optical characterization of ZnO nanoparticles capped withvarious surfactants. J. Lumin. 129, 434–438 (2009).

25. Tang, H. X. et al. Gas sensing behavior of polyvinylpyrrolidone-modified ZnOnanoparticles for trimethylamine. Sens. Actuat. B 113, 324–328 (2006).

26. Amb, C. M. et al. Dithienogermole as a fused electron donor in bulkheterojunection solar cells. J. Am. Chem. Soc. 133, 10062–10065 (2011).

27. Chu, T. Y. et al. Bulk heterojunction solar cells using thieno[3,4-c]pyrrole-4,6-dione and dithieno[3,2-b:2′ , 3′-d]silole copolymer with a power conversioneffciency of 7.3%. J. Am. Chem. Soc. 12, 4250–4253 (2011).

28. Subbiah, J., Kim, D. Y., Hartel, M. & So, F. MoO3/TFB double interlayer effecton polymer solar cells. Appl. Phys. Lett. 96, 063303 (2010).

29. Subbiah, J. et al. Combined effects of MoO3 interlayer and PC70BM on polymerphotovoltaic device performance. Org. Elect. 11, 955–958 (2010).

30. Kim, D. Y. et al. The effect of molybdenum oxide on organic solar cells. Appl.Phys. Lett. 95, 093304 (2009).

31. Irfan et al. Energy level evolution of molybdenum trioxide interlayer betweenindium tin oxide and organic semiconductor. Appl. Phys. Lett. 96,073304 (2010).

32. Meyer, J. et al. MoO3 films spin-coated from a nanoparticle suspension forefficient hole-injection in organic electronics. Adv. Mater. 23, 70–73 (2011).

33. Lilliedal, M. R. et al. The effect of post-processing treatments on inflection pointsin current–voltage curves of roll-to-roll processed polymer photovoltaics. Sol.Energy Mater. Sol. Cells 94, 2013–2031 (2010).

34. Aliaga, C. et al. Sum frequency generation and catalytic reaction studies of theremoval of organic capping agents from Pt nanoparticles by UV-ozonetreatment. J. Phys. Chem. C 113, 6150–6155 (2009).

35. Park, J. Y., Aliaga, C., Renzas, J. R., Lee, H. & Somorjai, G. A. The role of organiccapping layers of platinum nanoparticles in catalytic activity of CO oxidation.Catal. Lett. 129, 1–6 (2009).

36. Sun, Y. M., Seo, J. H., Takacs, C. J., Seifter, J. & Heeger, A. J. Inverted polymersolar cells integrated with a low-temperature-annealed sol–gel-derived ZnO filmas an electron transport layer. Adv. Mater. 23, 1679–1683 (2011).

37. Zhang, Y. T. et al. X-ray photoelectron spectroscopy study of ZnO films grownby metal-organic chemical vapor deposition. J. Crystal Growth 252,180–183 (2003).

AcknowledgementsThe authors acknowledge support from the Office of Naval Research (contract no.N000141110245) for the fabrication and characterization of polymer solar cells. F.S.acknowledges support from the Department of Energy Basic Energy Sciences (contract no.DE-FG0207ER46464) for the synthesis and characterization of ZnO composite films. J.R.R.acknowledges the support of the Air Force Office of Scientific Research (contract no.FA9550-09-1-0320) for the synthesis of the PDTG and PDTS polymers. The authors wouldalso like to thank the Major Analytical Instrumentation Center (MAIC) and the Dr AndrewRinzler research group at the University of Florida for their assistance in characterizing theZnO composite films.

Author contributionsC.E.S. performed polymer solar cell fabrication, characterization and data analysis, as wellas characterization of the ZnO–polymer films by AFM and XPS. S.C. contributed to dataanalysis, establishing a model to explain the ETL effects, and fabrication of the certifiedsolar cells. J.S. contributed to the early process optimization of the PDTG–TPD and PDTS–TPD BHJ solar cells. T-H.L. conceived the idea for modification of the ZnO–polymercomposite ETL by UV-ozone treatment and conducted optical transmissionmeasurements. C.A. synthesized the polymers used in this work. S-W.T. assisted inplanning and interpreting the data. F.S. and J.R.R. initiated and directed the researchproject. All authors discussed the results and commented on the manuscript.

Additional informationThe authors declare no competing financial interests. Supplementary informationaccompanies this paper at www.nature.com/naturephotonics. Reprints and permissioninformation is available online at http://www.nature.com/reprints. Correspondence andrequests for materials should be addressed to J.R.R. and F.S.





http://www.nature.com/reprints





high-efficiency inverted dithienogermole–thienopyrrolodione-based polymer solar cells

Documents