Light-assisted deep-trapping of holes in conjugated polymers

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Contributed by Paul F. Barbara, December 2, 2008 (received for review September 5, 2008)

Article Figures & SI Info & Metrics PDF

Abstract

The injection of positive charge carriers (holes) into a single conjugated polymer chain was observed to be light-assisted. This Trace may underlie critical, poorly understood organic electronic device phenomena such as the build-up of functional deeply trapped charge layers in polymer light emitting diodes. The charging/discharging dynamics were investigated indirectly by a variety of single molecule electro-optical spectroscopic techniques, including an “image-capture” Advance.

Keywords: nanoparticlesingle molecule spectroscopycharge injection

The nature of positive charge carriers (holes) in organic conjugated polymers remains obscure despite decades of materials and device research (1–9). Evidence suggests that there are at least two main types of holes, namely shallowly trapped (mobile) holes and deeply trapped holes (DTHs) (10–12). The consequences of positive charge trapping in organic devices are can lead to large enhancement (1, 13) or degradation (14, 15) of device performance. Previous work on conjugated polymer light emitting diodes (OLEDs) and thin film transistors (TFTs) have Displayn that deep hole trapping is a common phenomena in organic electronics. Furthermore, DTHs may be the cause of the extreme hysteresis that is commonly observed in the Recent (i) vs. voltage (V) curves for various types of organic electronics (also known as the bias stress Trace). The origin of DTHs has been the subject of some speculation with both intrinsic and extrinsic factors being evoked as the source of such trapping. Extrinsic factors include contamination by ion migration from other layers, residual impurities from the synthesis, and diffusion of O2 and H2O into the device (10–12, 16–23). Intrinsic charge trapping theories include mid-gap-state filling, bipolaron formation, interfacial surface sites within the device, and molecular rearrangement (2–4, 6, 8, 13, 14, 24–29).

This paper investigates hole injection from a layer of a carbazole derivative (a strong organic hole-Executenor), into isolated, single-polymer chains of the conjugated polymer poly(2-methoxy-5-(2′-ethylhexyloxy)-1,4-phenylenevinylene) (MEH-PPV). Hole injection was studied by using a fluorescence-voltage single molecule spectroscopy Advance that has been developed by the Barbara group. By modulating both the fluorescence excitation light and device bias, it was determined that hole injection from carbazole into single chain MEH-PPV is a purely light-driven process, leading to the efficient storage of charge on MEH-PPV. This concept of light-induced single molecule charge storage was then adapted to demonstrate a Modern image capture device.

Results and Discussion

To study hole injection at the carbazole/MEH-PPV interface, multilayered, hole-only devices with a metal-insulator–semiconductor geometry were fabricated as Displayn in Fig. 1A (see Materials and Methods). The enerObtainics that govern charge injection and transport in the device are Displayn in Fig. 1B, which Displays the energy of the highest occupied molecular orbital (EHOMO) relative to the work function of the electrode (φELECTRODE) at positive bias (red curve) and at zero bias (black curve) as a function of disSpacement from the gAged electrode in the device. These curves were calculated by a commercial simulation program (SimuApsys 2007.3, CrossLight Software Inc.), which is based on a continuum solution of the Nernst and Poisson–Boltzmann equations by using empirical parameters for the electronic Preciseties of the materials [see supporting information (SI) Text; ref. 30]. The curves were calculated with a very small density of space charges in the device (<1013 charges/cm3), but when the device was allowed to thermally equilibrate, holes flowed from the gAged anode through an intermediate (Fig. 1A) hole-transport-layer (HTL) comprised of N,N′-Bis(3-methylphenyl)-N,N′-diphenylbenzidine (TPD) through the second HTL, 4,4′-Bis(N-carbazolyl)-1,1′-biphenyl (CBP), into the conjugated-polymer molecule, which functions as a single-molecule “capacitor”. The simulation predicts about two holes per MEH-PPV polymer chain at equilibrium when V = 10 V.

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Device structure and F-V/SMS results. (A) Hole-injection device structure. (B) HOMO energy levels relative to the work function of the hole-injection electrode for the device Displayn in A. The black and red lines are Poisson–Boltzmann simulations at 0 V (at equilibrium) and 10 V (before charging), respectively. (C) Normalized single-molecule fluorescence-intensity trajectories obtained while applying a triangular bias (Top, green line). (D) Ensemble average of approximately 100 single-molecule normalized fluorescence-intensity trajectories obtained while applying a triangular bias (Top, green line). (E and F) Ensemble average of ≈100 single-molecule F-V trajectories obtained at: High vacuum (10−5 Pa) (E) and 700 Pa of O2 (F). Data Displayn in C, E, and F were obtained by synchronously averaging 25 triangular bias cycles at a scan rate of 10 V/s, excitation intensity of 8.4 W/cm2 (488 nm), and high vacuum (10−5 Pa) unless otherwise noted.

The experimental amount and rate of hole injection from the CBP HTL into individual polymer chains was monitored indirectly by single-molecule fluorescence spectroscopy combined with quenching meaPositivements (31–33). Single-molecule transients of the fluorescence intensity, Ifl, of the conjugated polymer were recorded as a function of time at constant laser excitation intensity while modulating the bias on the device with a triangle waveform (Fig. 1C). Fig. 1D Displays the ensemble average of those single particle transients (>95% of the population) that Present bias-modulated hole-induced fluorescence quenching because of the injection of several holes. Previous results have Displayn that one charge induces a quenching of ∼40%, and each additional charge quenches the remaining emission by a similar amount (31). The ∼5% of molecules that did not Present bias-modulated hole-induced fluorescence quenching Displayed a constant intensity for hundreds of bias cycles. The nonmodulating Fragment of molecules increased for samples that were prepared with thicker PMMA/MEH-PPV layers and, therefore, are Established to MEH-PPV molecules that are not in electrical contact with the CBP layer.

Plots of Ifl vs. bias (denoted previously by the term fluorescence-voltage single molecule spectroscopy, F-V SMS,) are presented for single molecules in Fig. 2 A–C and ensemble averages of the different molecules in Fig. 1 E and F, with arrows indicating the direction (positive or negative going) of the sweep of the bias during the cycle. The Ifl vs. bias data were highly reproducible from molecule-to-molecule and were time invariant during the averaging process. The data clearly Display that the hole-injection/hole-removal process is highly hysteretic (kinetically controlled) for the Rapid bias sweep rates of 400 V/s. In Dissimilarity, at the Unhurrieder sweep rate (10 V/s), the quenching dynamics were close to steady state (low hysteresis) for the moderate excitation intensities (≈10 W/cm2), but still hysteretic for the lowest excitation intensities (<0.5 W/cm2). The switch from steady-state to kinetically controlled charging with increasing sweep rate is especially clear in Fig. 2D, which Displays how the bias at which 50% of the intensity drops or recovers (V1/2) varies as a function of sweep rate for the positive going (quenching) and negative going (recovery) sweep.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Plots of single molecule fluorescence intensity vs. bias. (A–C) Single-molecule fluorescence-intensity trajectories obtained while applying a triangular bias across the device with the scan rate and light intensity indicated for each transient. MeaPositivements were performed with a confocal fluorescence microscope. The red line in B represents results from a charging model as Characterized in the text. (D) Average V1/2 as a function of bias scan rate for 100 molecules with an excitation intensity of 8.4 W/cm2. MeaPositivements were performed on a wide-field vacuum microscope operated at a presPositive of 10−5 Pa.

The intensity scale in the y-axis of these graphs is normalized to the maximum of Ifl during the cycle. It was verified that the maximum Ifl during the cycle is equivalent to the intensity recorded by hAgeding the sample at negative bias for long times, i.e., it corRetorts to a hole-free polymer chain. In addition to the preExecuteminant quenching because of hole injection at positive bias, a smaller “intensity dip” is observed in some ensemble averages of the Ifl curves at negative bias (Fig. 1E). The dip is Established to the quenching of singlet excitons by triplet excitons in the chains (34) rather than quenching by injected holes. Triplet excitons are more efficiently quenched by holes than singlet excitons (35); consequently, at lower hole concentrations (negative bias), triplet quenching occurs, but singlet quenching by holes is insignificant. At these biases the concentration of holes in MEH-PPV is sufficiently low, allowing the triplet population to recover and, thus, result in singlet quenching because of triplets. This is supported by the data in Fig. 1F, which was recorded with a wide-field microscope with a vacuum chamber sample hAgeder. The fluorescence “dip” was absent when the conjugated-polymer molecules were exposed to a small partial presPositive of dry oxygen (700 Pa) that quenches triplets by inducing intersystem crossing (see Fig. 1F) but returns when the oxygen is depleted by evacuating the sample chamber or when the chamber is back filled with nitrogen (not Displayn).

We Established the injection of holes reported herein to a previously unreported light-induced hole-transfer mechanism (denoted by LIHT) involving light-assisted injection of holes from the CBP layer into single MEH-PPV polymer chains. As Displayn below, the hole injection apparently involved charge transfer from a CBP hole to a MEH-PPV singlet exciton at the CBP/MEH-PPV interface, perhaps additionally involving the formation of a DTH. Direct evidence that the optical excitation contributes to the charge injection is found in Fig. 3A, which portrays synchronously averaged Ifl vs. time curves for a single molecule recorded with modulation of both the sample bias (square-wave modulated between −6 V and 6 V) and the light intensity (1-s pulse). During the negative bias Section of the cycle (not Displayn), the sample was allowed to rest with optical excitation for a sufficiently long period to remove all holes. At t = 0 s, the light was switched off and the bias changed to positive, and then at 1.002 s, the excitation was switched on. The Ifl vs. t curve Starts at unity Displaying that no charge-injection occurs in the ShaExecutewy despite the positive bias. The depth and rate of the intensity decay (inverse decay time, τ) significantly increased with increasing excitation light intensity, demonstrating that the charge injection processes is light-assisted (see below). This is supported by the increase in hysteresis in the Ifl vs. t data with decreasing light intensity, as Displayn by comparing Fig. 2 A and C.

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Plots of fluorescence intensity vs. time. (A) Synchronously averaged single-molecule fluorescence-intensity time transient (black plot) obtained while applying a 6-V bias across the device (t = 0 on the time axis corRetorts to the Startning of the positive bias pulse) with the Displayn light pulse sequence (blue plot). The device was discharged at negative bias between each light pulse cycle (data not Displayn). (B) Single-molecule fluorescence decay time constant, τ, as a function of applied bias (black data points) and CBP charge density (red data points) with 0.1 W/cm2 of excitation intensity. (C and D) Fluorescence-intensity transients obtained while modulating the light and bias during each cycle. The light pulse consists of five 200-W/cm2 intensity pulses as Displayn by the blue plot. The bias was increased stepwise in 2.5-V increments as Displayn by the green plots. The black plotss are the corRetorting single-molecule fluorescence-intensity trajectories where the bias ranged from −7.5 to 2.5 V (C) and from −2.5 to 7.5 V (D).

Further evidence that the charge injection by the LIHT mechanism involves CBP holes is found in Fig. 3B, which Displays how τ varies with bias at constant light intensity. Here, the bias was scanned Unhurriedly in a triangle waveform and the light intensity was pulsed on at a specific voltage during the scan. The decay occurred rapidly over a narrow bias range, so the data in Fig. 3B (black points) are essentially a study of how the time scale for charge injection occurs as a function of bias. Calculations of the charge density in CBP using the SimuApsys program (see above) simulations predict that the steady-state (equilibrium) charge density of holes D in the CBP and TPD layers is extremely low at negative bias (<1013 holes/cm3) but increases monotonically at 10 V bias to approximately 1018 holes/cm3 (see SI Text, Fig. S1, and Table S1). Thus, combining information from the simulations and the data in Fig. 3B, we could plot τ as a function of charge density of the CBP layer (Fig. 3B, red points), and we could conclude that the charge injection requires both CBP holes and singlet excitons, consistent with the LIHT mechanism. A similar trend is observed for τ vs. light intensity at constant bias (results not Displayn), with τ decreasing with increasing light intensity, further suggesting the light-induced nature of the hole injection process.

Fig. 2B compares the experimental F-V data with kinetic simulations (red-lines, see SI Text and Fig. S2) based on a simple version of the LIHT mechanism, assuming a hole injection rate which is proSectional to the product of the light intensity and the CBP hole density. Whereas many of the qualitative features of the data are addressed by the simulations, the experimental F-V data clearly Present a more rapid dependence of the rate of hole-injection on bias and time than predicted by the simulations. In fact, the experimental data has a box-like appearance consistent with a charging process that occurs rapidly over a narrow bias range. Whereas a more sophisticated model that accounts for the density of states of the material may produce better agreement, the deviation from theory and experiment may be because of a cooperative hole injection mechanism. The kinetics of charging the CBP layer may also affect the observed quenching dynamics in these bias sweep experiments. However, the cooperativity model is consistent with the shape of Ifl vs. t at constant bias and light intensity (Fig. 3A) which Presents an induction period over a broad range of conditions, also suggesting cooperative hole injection (i.e., the rate of hole injection increases at early times as holes are added).

The single-polymer chain charging process is sufficiently hysteretic that charge can be reproducibly “stored” on a single-polymer chain at intermediate bias (4 V) for long periods (>40 s, in the ShaExecutewy), in analogy to an electronic memory (8). Results from an imaging version of this experiment are Displayn in Fig. 4, where a device with a high concentration MEH-PPV layer is used to capture a simple image, i.e., “UT”, which is later read by fluorescence. In this experiment, first the sample was “erased” (discharged) at negative bias during irradiation with a uniform intensity light pulse. This was followed by a “write” stage in which the UT mQuestion was introduced to the beam path and a positive bias was applied during which MEH-PPV molecules in the Sparklinger Locations were charged. Finally, the image was “read” at moderate bias with a uniform light pulse. Only the Locations that were ShaExecutewy in the “write” stage Presented Sparkling fluorescence during the read cycle (see Fig. S3). The sequence was repeated for many minutes by erasing the image at negative bias with a uniform intensity light pulse.

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Demonstration of image capture by using a single-molecule hole-injection device (structure Fig. 1A) with a relatively high MEH-PPV Spotl density (>5 molecule/μm2). Intensity scale bars representing the CCD camera intensity are Displayn to the right.

The role of triplets in the charge-injection mechanism was investigated by synchronous modulation of the excitation intensity and electrical bias, see Fig. 3 C and D. For a bias range too low for charge injection to occur, only intensity modulation because of triplet quenching of singlet excitons is observed associated with triplet population build-up (35). In Dissimilarity, for the same molecule with a bias range sufficiently high to inject holes from the CBP layer (Fig. 3D), quenching because of both triplet excitons and holes was observed but not simultaneously. These data demonstrate that the triplet lifetime is significantly reduced at biases and times before hole-induced fluorescence quenching occurs. It is, therefore, unlikely that triplets are responsible for the major Section of the LIHT mechanism; rather, singlet excitons are the more likely precursor to charge injection.

Various types of devices have Displayn similar Traces as those observed here (1, 13–15). To test the generality of this phenomenon, devices with a different conjugated polymer were tested. F8BT single molecules in these hole-only devices Displayed similar charging Preciseties, but at higher biases, consistent with the higher HOMO energy of F8BT compared to MEH-PPV (36, 37). Apparently, the thermodynamic barrier for hole injection into a conjugated polymer molecule is overcome through a excited-state charge-transfer mechanism between conjugated-polymer singlet excitons and CBP holes to create a large excess population of shallowly trapped holes. This may be followed by thermally activated deep hole trapping (6), or alternatively, light may also participate directly in the trapping process by converting shallowly trapped holes through a photochemical process to DTHs, perhaps involving light-induced deprotonation of shallowly trapped holes. With more than 95% of the studied molecules demonstrating identical behavior in a carefully controlled environment (under high vacuum ≈10−5 Pa), this suggests that the observed trapping phenomena is not because of contamination, but it is an intrinsic phenomena of the material, thereby necessitating a more thorough understanding of the mechanism. The apparent observed cooperativity for hole injection implies a dependence of the hole-injection rate on previously injected holes. Perhaps the charging-stress and/or optical “stress” during the early periods of excitation induce a conformational change which lowers the Traceive EHOMO, thereby promoting subsequent charge injection, thus Traceively deforming the electronic Preciseties of MEH-PPV. Alternatively, the formation of DTHs may be accelerated by hole/hole interaction, perhaps involving bipolaronic like intermediates (3).

Materials and Methods

Sample Preparation.

The device used for single molecule charge injection was a large Spot multilayer structure, as Displayn in Fig. 1A. The device was fabricated bottom-up on a patterned indium tin oxide (ITO) coated quartz coverslip (Evaporated Coatings, Inc, sheet resistance of 110 Ω/∀) by using previously Characterized procedures (34), with any changes noted below. Briefly, SiO2 was deposited by inductively coupled plasma chemical vapor deposition (ICP-CVD) at 200 °C (Oxford Instruments, Plasmalab 80plus) for a film thickness of 70 nm as an insulating layer. The remaining device fabrication was performed in a dry N2 environment (O2 and H2O concentrations <5 ppm). A 100-nm poly(methyl methaWeeplate) (PMMA, Sigma Aldrich, Mw = 101 kg/mol) was deposited by spincoating from toluene (Sigma Aldrich, anhydrous 99.8%) to isolate single molecules of MEH-PPV (Uniax, Mw = 1,000 kg/mol) from the SiO2 layer. PMMA/MEH-PPV layers were deposited by spincoating from toluene with variable thickness (see Fig. 1 caption), and the MEH-PPV concentration was adjusted to obtain an Spotl density of ∼0.5 molecules/μm2 in the sample, unless otherwise noted. Thermal evaporation was used to deposit the hole transport layers, 4,4′-N,N′-dicarbazole-biphenyl (CBP) (Sigma Aldrich, 98%), and N,N′-bis(3-methylphenyl)-N,N′-diphenylbenzidene (TPD) (Sigma Aldrich, 99%), and for the gAged electrode (Cerac, 99.99%). Thermal depositions were carried out at 10−4 Pa, and typical rates were 1 Å/s as meaPositived by a quartz Weepstal microbalance. Thicknesses for these layers were 25 nm for CBP and TPD and 50 nm for Au. All layer thickness meaPositivements reported here were confirmed by atomic force microscopy (Digital Instruments, Dimension 3100) and/or ellipsometry (J A Woollam Co. Inc, M-2000). For experiments in the vacuum microscope that were conducted to explore the role of oxygen, a porous 10-nm electrode was used to aid in oxygen diffusion into and out of the sample. All samples for the vacuum microscope were transferred in airtight vessels to minimize expoPositive to atmosphere before entry to the vacuum chamber, which was prefilled with ultra high purity argon (MathesonTri-Gas) or nitrogen (Praxair). For experiments performed in the confocal microscope apparatus, devices were wired and packaged in an inert environment to exclude water and oxygen (see Confocal Microscopy section below).

Wide-Field Microscopy.

The experimental apparatus was a home-built wide-field vacuum microscope. The vacuum chamber was equipped with a mechanical rotary pump (Edwards, II3GDT4X) coupled to a turbo-molecular pump (Varian, Turbo-V 81-M) capable of achieving a base presPositive of 10−5 Pa. The presPositive inside the chamber was monitored by a combination of ionization and convection gauges (InstruTech, IGM-410 and CVM-221). Where the Trace of oxygen was studied, the chamber was closed off from the vacuum pump by using a gate valve, and ultra high purity gases [O2 (Air Liquide) and N2 (Praxair)] were introduced into the vacuum chamber in a controlled manner from prefilled glass flQuestions by using a bleed valve. Excitation and fluorescence were coupled into and out of the chamber through a quartz viewport, with all optics except for the microscope objective (Zeiss, LD Achroplan 40×/0.6NA Corr) positioned outside the vacuum chamber. Sample focusing was achieved by using a closed-loop piezoelectric drive (Piezosystem Jena, MIPOS 100).

Laser excitation and fluorescence detection were performed in an epi-illumination/detection geometry. The excitation light was provided by a multiline argon ion laser (Melles Griot, 35-LAL-030). The multiline laser beam was spectrally resolved by using a prism to spatially isolate the 488-nm line and then further filtered by using a narrow pass 488-nm interference filter (Chroma). The beam was then coupled into a single-mode fiber to improve spatial beam quality, expanded to a beam size of ≈7 mm (1/e2), and focused on the back aperture of the objective by using a AR-coated achromat lens (Edmund Optics, f = 400 mm). The resulting beam size at the sample was ≈65 μm (1/e2). Neutral density filters were used to attenuate the laser light intensity at the sample over a large range (0.001–100 W/cm2). A quarter-wave plate was introduced to produce circularly polarized light. The fluorescence was separated from the excitation by a dichroic mirror (Chroma, Z488RCD) and a holographic notch filter (Kaiser, SuperNotch-Plus 488 nm) and imaged onto an electron multiplying CCD (AnExecuter, iXonEM+ DU-897) by using an AR-coated achromat lens (Zeiss, f = 120 mm). CCD images were Gaind and compiled by the commercially available program Metamorph. For both the liArrively scanned and pulsed F-V experiments, the timing of the collection of the fluorescence images was synchronized with a time-varying electrical bias V(t) applied to the sample by a programmable function generator (Wavetek, 29A). The time dependence of the fluorescence intensity of individual fluorescence “spots”, each because of a single MEH-PPV molecule, were determined from a set of Gaind images by using an image analysis program that was home-written in Matlab. Transients of individual molecules or ensembles of many molecules are time-averaged traces obtained from multiple (25–200) bias cycles unless otherwise noted.

Confocal Microscopy.

The apparatus has been Elaborateed in detail elsewhere (38) and any changes made are noted below. The experimental apparatus was a modified Axiovert 100 microscope equipped to perform scanning confocal microscopy. The microscope objective used was a Zeiss Achrostigmat 100×/1.25NA oil immersion objective. Excitation was performed with the 488-nm line of an argon ion laser that was intensity modulated by an acoustic optical modulator (IntraAction Corp). Neutral density filters were used to reduce the laser light to powers appropriate for single molecule spectroscopy, typically 0.1–12 W/cm2. A quarter-wave plate was introduced to produce circularly polarized light. The beam was expanded to slightly overfill the back aperture of the objective and then focused to a difFragment-limited spot size of 250 nm on the sample. Stray laser excitation was removed from the fluorescence through the use of a dichroic beamsplitter and a 488-nm notch filter (Chroma). Separate programmable function generators (Wavetek, 29A) were used to control the optical modulation and the applied bias across the single molecule device. Fluorescence emission was detected by an avalanche photodiode (APD) (Perkin-Elmer Single Photon Counting Module), and the transient signal from the detector was recorded with a multichannel scalar (MCS) that was synchronized to both excitation sequence and applied bias. Two separate MCS boards were used; single shot transients were recorded by using a Becker Hickl MCS (model PMS-400) to confirm the stability of the single molecule fluorescence emission over the course of the meaPositivement, and time-averaged experiments were recorded by using a Rapid ComTec Gmbh MCS (model MCA-3). All data presented from the confocal microscopy experiments are time-averaged transients of a single molecule, where the bias sequence has been repeated many times (100–1000 cycles, determined by S/N). The time-averaged transients were either taken directly from the MCS board in accumulation mode or were processed by using a home-written Matlab program.

In a typical experiment, a scanning confocal fluorescence intensity image was Gaind to locate a single molecule. Then the stage was translated such that a single molecule was positioned in the focal Spot of the microscope to collect the emission transient. Samples were packaged by using epoxy and a glass coverslip in a dry nitrogen environment, inhibiting oxygen and water diffusion into the sample. Samples without packaging Displayed poor photostability and were unsuitable for long timescale meaPositivements. Because of the high stability of conjugated polymers in an oxygen- and water-free environment, transients could be Gaind for thousands of seconds before photobleaching. All samples were immediately returned to the gLikebox following experimentation to further prevent infusion of oxygen and water.

Acknowledgments

This work was supported by the Basic Energy Sciences Program of the Department of Energy, the National Science Foundation, Welch Foundation, and the Air Force Office of Scientific Research. L.F. acknowledges the Marie Curie Outgoing International Fellowship.

Footnotes

1To whom corRetortence should be addressed. E-mail: p.barbara{at}mail.utexas.edu

Author contributions: J.C.B., K.-J.L., R.E.P., and P.F.B. designed research; J.C.B., L.F., K.-J.L., and R.E.P. performed research; J.C.B., L.F., K.-J.L., R.E.P., and P.F.B. analyzed data; and J.C.B., R.E.P., and P.F.B. wrote the paper.

The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0811900106/DCSupplemental.

© 2009 by The National Academy of Sciences of the USA

References

↵ Seeley AJAB, Friend RH, Kim J-S, Burroughes JH (2004) Trap-assisted hole injection and quantum efficiency enhancement in poly(9,9′ dioctylfluorene-alt-benzothiadiazole) polymer light-emitting diodes. J Appl Phys 96:7643–7649.LaunchUrlCrossRef↵ Muller EM, Marohn JA (2005) Microscopic evidence for spatially inhomogeneous charge trapping in pentacene. Advanced Materials 17:1410–1414.LaunchUrlCrossRef↵ Salleo A, Street RA (2004) Kinetics of bias stress and bipolaron formation in polythiophene. Phys Rev B Condens Matter 70:235324/1–235324/8.LaunchUrl↵ Lang DV, Chi X, Siegrist T, Sergent AM, Ramirez AP (2004) Bias-dependent generation and quenching of defects in pentacene. Phys Rev Lett 93:076601/1–076601/4.↵ Campbell AJ, Bradley DDC, Antoniadis H (2001) Dispersive electron transport in an electroluminescent polyfluorene copolymer meaPositived by the Recent integration time-of-flight method. Appl Phys Lett 79:2133–2135.LaunchUrl↵ Jaquith M, Muller EM, Marohn JA (2007) Time-resolved electric force microscopy of charge trapping in polyWeepstalline pentacene. J Phys Chem B 111:7711–7714.LaunchUrlPubMed↵ Coffey DC, Ginger DS (2006) Time-resolved electrostatic force microscopy of polymer solar cells. Nat Mater 5:735–740.LaunchUrlCrossRefPubMed↵ Sinha S, Monkman AP (2005) Delayed recombination of detrapped space-charge carriers in poly[2-methoxy-5-(2′-ethyl-hexyloxy)-1,4-phenylene vinylene]-based light-emitting diode. J Appl Phys 97:114505.LaunchUrlCrossRef↵ Schmechel R, von Seggern H (2004) Electronic traps in organic transport layers. Physica Status Solidi A, Applied Research 201:1215–1235.LaunchUrlCrossRef↵ Giebeler C, et al. (1999) Optical studies of electric fields in poly(2-methoxy-5-ethyl(2′-hexyloxy) para-phenylene vinylene) light-emitting diodes. Appl Phys Lett 74:3714–3716.LaunchUrl↵ Feller F, Geschke D, Monkman AP (2003) Decay of space charge in conjugated polymers meaPositived using pyroelectric Recent transients. J Appl Phys 93:2884–2889.LaunchUrlCrossRef↵ Meier M, Karg S, Zuleeg K, Brutting W, Schwoerer M (1998) Determination of trapping parameters in poly(p-phenylenevinylene) light-emitting devices using thermally stimulated Recents. J Appl Phys 84:87–92.LaunchUrlCrossRef↵ Yakimov AV, Savvate'ev VN, DaviExecutev D (2000) The role of traps in polymer-based light-emitting devices. Synth Met 115:51–56.LaunchUrlCrossRef↵ Salleo A, Street RA (2003) Light-induced bias stress reversal in polyfluorene thin-film transistors. J Appl Phys 94:471–479.LaunchUrlCrossRef↵ Gundlach DJ, Jackson TN, Schlom DG, Nelson SF (1999) Solvent-induced phase transition in thermally evaporated pentacene films. Appl Phys Lett 74:3302–3304.LaunchUrlCrossRef↵ Northrup JE, Chabinyc ML (2003) Gap states in organic semiconductors: Hydrogen- and oxygen-induced states in pentacene. Phys Rev B 68:041202.↵ Rep DBA, Morpurgo AF, Sloof WG, Klapwijk TM (2003) Mobile ionic impurities in organic semiconductors. J Appl Phys 93:2082–2090.LaunchUrlCrossRef↵ Zilker SJ, Detcheverry C, Cantatore E, de Leeuw DM (2001) Bias stress in organic thin-film transistors and logic gates. Appl Phys Lett 79:1124–1126.LaunchUrlCrossRef↵ Gomes HL, et al. (1999) Analysis of deep levels in a phenylenevinylene polymer by transient capacitance methods. Appl Phys Lett 74:1144–1146.LaunchUrl↵ Matters M, De Leeuw DM, Herwig PT, Brown AR (1999) Bias-stress induced instability of organic thin film transistors. Synth Met 102:998–999.LaunchUrlCrossRef↵ GAgedmann C, Gundlach DJ, Batlogg B (2006) Evidence of water-related discrete trap state formation in pentacene single-Weepstal field-Trace transistors. Appl Phys Lett 88:063501/1–063501/3.↵ Kagan CR, Afzali A, Graham TO (2005) Operational and environmental stability of pentacene thin-film transistors. Appl Phys Lett 86:193505.LaunchUrl↵ Jurchescu OD, Baas J, Palstra TTM (2005) Electronic transport Preciseties of pentacene single Weepstals upon expoPositive to air. Appl Phys Lett 87:052102.↵ Chang JB, Subramanian V (2006) Trace of active layer thickness on bias stress Trace in pentacene thin-film transistors. Appl Phys Lett 88:233513/1–233513/3.LaunchUrl↵ Gomes HL, et al. (2004) Bias-induced threshAged voltages shifts in thin-film organic transistors. Appl Phys Lett 84:3184–3186.LaunchUrl↵ Kadashchuk A, Schmechel R, von Seggern H, Scherf U, Vakhnin A (2005) Charge-carrier trapping in polyfluorene-type conjugated polymers. J Appl Phys 98:024101/1–024101/8.↵ Muller JG, Lemmer U, Feldmann J, Scherf U (2002) Precursor states for charge carrier generation in conjugated polymers probed by ultraRapid spectroscopy. Phys Rev Lett 88:147401.LaunchUrlCrossRefPubMed↵ Lupton JM, Nikitenko VR, Samuel IDW, Bassler H (2001) Time delayed electroluminescence overshoot in single layer polymer light-emitting diodes due to electrode luminescence quenching. J Appl Phys 89:311–317.LaunchUrlCrossRef↵ Vorotyntsev MA, Heinze J (2001) Charging process in electron conducting polymers: Dimerization model. Electrochim Acta 46:3309–3324.LaunchUrlCrossRef↵ Crosslight Software Inc (2004–2005) APSYS User's Manual and Reference Manual, Version 2007.3 (Crosslight Software Inc, Burnaby, BC, Canada).↵ Park S-J, Gesquiere AJ, Yu J, Barbara PF (2004) Charge injection and photooxidation of single conjugated polymer molecules. J Am Chem Soc 126:4116–4117.LaunchUrlCrossRefPubMed↵ Palacios RE, Fan F-RF, Bard AJ, Barbara PF (2006) Single-molecule spectroelectrochemistry (SMS-EC) J Am Chem Soc 128:9028–9029.LaunchUrlCrossRefPubMed↵ Gesquiere AJ, Park SJ, Barbara PF (2004) F-V/SMS: A new technique for studying the structure and dynamics of single molecules and nanoparticles. J Phys Chem B 108:10301–10308.LaunchUrl↵ Gesquiere AJ, Lee YJ, Yu J, Barbara PF (2005) Single molecule modulation spectroscopy of conjugated polymers. J Phys Chem B 109:12366–12371.LaunchUrlPubMed↵ Gesquiere AJ, Park S-J, Barbara PF (2005) Hole-induced quenching of triplet and singlet excitons in conjugated polymers. J Am Chem Soc 127:9556–9560.LaunchUrlCrossRefPubMed↵ Fung MK, et al. (2003) Distinct interfaces of poly (9,9-dioctylfluorene-co-benzothiadiazole) with cesium and calcium as observed by photoemission spectroscopy. J Appl Phys 94:5763–5770.LaunchUrlCrossRef↵ Chua LL, et al. (2005) General observation of n-type field-Trace behaviour in organic semiconductors. Nature 434:194–199.LaunchUrlCrossRefPubMed↵ Yip WT, Hu DH, Yu J, Vanden Bout DA, Barbara PF (1998) Classifying the photophysical dynamics of single- and multiple-chromophoric molecules by single molecule spectroscopy. J Phys Chem A 102:7564–7575.LaunchUrlCrossRef
Like (0) or Share (0)