Probing the Energy Level Alignment and the Correlation with Open

Subscriber access provided by UNIV OSNABRUECK

Article

Probing the Energy Level Alignment and the Correlation with Open-Circuit Voltage in Solution Processed Polymeric Bulk Heterojunctions Photovoltaic Devices Qing-Dan Yang, Ho-Wa Li, Yuanhang Cheng, Zhiqiang Guan, Taili Liu, Tsz-Wai Ng, Chun-Sing Lee, and Sai Wing Tsang ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b11395 • Publication Date (Web): 01 Mar 2016 Downloaded from http://pubs.acs.org on March 2, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Probing the Energy Level Alignment and the Correlation with Open-Circuit Voltage in Solution Processed

Polymeric

Bulk

Heterojunctions

Photovoltaic Devices Qing-Dan Yang1,2‡, Ho-Wa Li1‡, Yuanhang Cheng1, Zhiqiang Guan1,2, Taili Liu1, Tsz-Wai Ng1,2, Chun-Sing Lee1,2*, Sai-Wing Tsang1* 1. Department of Physics and Materials Science, City University of Hong Kong, Hong Kong SAR, P. R. China. 2. Department of Physics and Materials Science, Center of Super-Diamond and Advanced Films (COSDAF), City University of Hong Kong, Hong Kong SAR, P. R. China. KEYWORDS: photoemission spectroscopy, energy diagram, bulk heterojunction, photovoltage loss, photovoltaics

1 Environment ACS Paragon Plus


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Energy level alignment at organic donor and acceptor interface is a key to determine the photovoltaic performance in organic solar cells, but direct probing such energy alignment is still challenging especially for solution processed bulk heterojunction (BHJ) thin films. Here we report a systematic investigation on probing the energy level alignment with different approaches in five commonly used polymer: [6,6]-phenyl-C71-butyric acid methyl ester (PCBM) BHJ systems. We find that by tuning the weight ratio of polymer to PCBM, the electronic features from both polymer and PCBM can be obtained by photoemission spectroscopy. Using this approach, we find that some of the BHJ blends simply follow vacuum level alignment, but others show strong energy level shifting as a result of Fermi level pinning. Independently, by measuring the temperature dependent open-circuit voltage (VOC), we find that the effective energy gap (Eeff), the energy difference between the highest occupied molecular orbital of the polymer donor (EHOMO-D) and lowest unoccupied molecular orbital of the PCBM acceptor (ELUMO-A), obtained by photoemission spectroscopy in all polymer:PCBM blends has an excellent agreement with the extrapolated VOC at 0 K. Consequently, the photovoltage loss of various organic BHJ photovoltaic devices at room temperature is in a range of 0.3 V to 0.6 V. It is believed that the demonstrated direct measurement approach of the energy level alignment in solution processed organic BHJ will bring deeper insight into the origin of the VOC and the corresponding photovoltage loss mechanism in organic photovoltaic cells.


Page 2 of 33

Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION Since the first demonstration of efficient organic light emitting diode (OLED) and organic photovoltaic (OPV) cell with the application of organic heterojunctions in 1980s,1,2 tremendous research effort has been devoted to synthesizing new materials and engineering interfacial properties to improve the efficiency of organic electronic devices.3-9 Over the years, it has been well known that the energy level alignment and charge transfer (CT) states at the heterojunctions play crucial roles in controlling charge injection and exciton dissociation efficiencies.10-15 Particularly in OPVs, energy level alignment at the heterojunction has been the rule-of-thumb for designing new materials to improve the open-circuit voltage (VOC) in devices.16-19 Scharber et al. has proposed that the VOC of organic heterojunction photovoltaic cell is proportional to the energy difference between the highest-occupied molecular-orbital (EHOMOD)

of the electron-donor material and the lowest-unoccupied molecular-orbital (ELUMO-A) of the

electron-acceptor material, which is later been regarded as the effective energy gap (Eeff) of organic heterojunctions.20 In most cases, the reported Eeff is simply determined by assuming a common vacuum level alignment at the organic donor and acceptor interface. However, it has been widely reported that charge transfer between the donor and acceptor materials would introduce interface dipole and/or energy level bending. For instance, 0.4 eV to 0.6 eV interface dipole at organic-organic interface have been reported in both conjugated small molecules and polymers when they are in contact with fullerenes.21,22 Also, considerable energy level bending as large as 0.5 eV to 0.8 eV have been reported at various organic-organic interfaces.23-25 Given that the VOC of OPVs is typically less than 1 V, such large energy level re-arrangement when two materials in contact would severely introduce error for interpreting the physics of corresponding photovoltage losses behind. Therefore, those examples unambiguously suggest the need of



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

taking the energy level alignment in actual blended thin films in order to better understand the correlation between the VOC and Eeff. Recently, Widmer et al.26 investigated a series of small molecule OPV devices, they found that the maximum attainable VOC (VOC-MAX) obtained by extrapolating the temperature dependent VOC to 0 K is in well consistent with Eeff. Nevertheless, it is still in hot debate on whether the maximum value of VOC should solely depend on the energy level alignment or the additional charge transfer (CT) states induced at the heterojunctions.27-33 Undoubtedly, solving this important but controversial issue requires reliable experimental approach to accurately determinate the energy level alignment at organic heterojunctions. To date, several advanced approaches have been developed to investigate the energy level alignment, such as cross-sectional scanning Kelvin probe microscopy (SKPM)34 and synchrotron radiation photoemission spectroscopy.35 Still, the most reliable and widely adopted approach to determine the energy level alignment at organic heterojunction thin films is by photoemission spectroscopy (PES). It directly probes the binding and valence energy levels of electrons in solid-state thin films. This technique has been proved to be a viable tool to study the heterojunctions in small molecule based organic thin films, where well defined interface can be precisely controlled by thermal evaporation. However, it is well known that using PES is challenging to investigate solution processed bulk heterojunction (BHJ) films where the organic donor and acceptor materials are blended together in form of nano-scale inter-penetrating networks. Owing to non-uniform vertical phase segregation of organic donor and acceptor materials, it is difficult to get the electronic signals from both of them in a single survey. This hinders the information of any possible interactions between the donor and acceptor materials to be measured at the interface. Consequently, only a handful of examples were reported using such surface sensitive PES technique to investigate the energy level alignment in solution processed


Page 4 of 33

Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


BHJ organic photovoltaic thin films. For instance, Xu et al.37 reported that using ultra-violet photoemission spectroscopy (UPS) to study the energy level alignment of poly(3hexylthiophene) and [6,6]-phenyl-C71-butyric acid methyl ester (P3HT:PCBM) formed BHJ thin films. As the surface of the film was mostly covered by a thin layer of P3HT, no structural features from PCBM could be obtained.37 Alternatively, Guan et al.38 used a lift-off approach to turn the P3HT:PCBM blend film upside down on a second substrate and investigated the corresponding electronic structure by UPS and inverse photoemission spectroscopy (IPES), and successfully obtained the Eeff of 1.46 eV. It should be noted that this value is very different from the value (0.8 eV) obtained by taking the individual P3HT and PCBM energy levels based on the assumption of vacuum level alignment.38-41 Therefore, a facile approach which can directly measure the Eeff in organic BHJ thin films is of the utmost importance to understand both the interfacial electronic process and photovoltaic device operation. Here we report a systematic investigation on the approach of probing the energy level alignment in five commonly used solution processed polymer:fullerene BHJ blends. Due to the non-uniform vertical phase segregation of the BHJ films, we have found that an abnormally high (> 1:20) polymer:fullerene weight ratio has to be used in order to obtain the electronic features from both materials in a single survey by PES. Some of the BHJs simply follow the vacuum level alignment, but others show strong energy level shifting as a result of Fermi level pinning. In order to verify the Eeff extracted by using such high ratio on the corresponding photovoltaic devices, we compared the PES results with the photovoltaic devices fabricated in different ratios and the extrapolated VOC at 0 K. We have found that both PES and photovoltaic results are in excellent agreement. This not only provides a facile alternative to determine the Eeff at organic heterojunction, but also sheds light on the origin of the VOC in photovoltaic devices. Based on the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

above results, the photovoltage loss in solution processed BHJ devices and the impact of substrate effect on the energy level alignment are discussed. 2. RESULTS AND DISCUSSION Due to the difference of surface energies between polymer and PCBM, the polymer tends to accumulate on the surface of BHJ thin films. Since, there is a limited detection depth, 10 nm, by PES, it is challenging to probe the interfacial interaction between polymer and PCBM in BHJ thin films.35 Therefore, we sought to significantly increase the PCBM weight ratio in blend films until the HOMO feature peaks from both the polymer and PCBM could be obtained in a single scan. Figure 1 shows the contact angles of deionized (DI) water on (a) P3HT:PCBM and (b) PTB7:PCBM films with different weight ratios. Owing to the difference in surface energy, there is a notable variation in contact angles between the pure polymer and the pure PCBM. The measured contact angles for pure P3HT, PTB7 and PCBM are 108o, 107o, and 82o, respectively, and these are similar to those previously reported values.42-44 It is worth noting that, as shown in Figure 1c, for both polymer:PCBM blend films, the contact angle is only slightly decreased by less than 10o when the weight ratio is increased from 1:0 to 1:10 for both P3HT:PCBM and PTB7:PCBM blend films. This small change in contact angle suggests that the surface of the blend film is still dominated by the polymer even at such high weight ratio of 1:10. Interestingly, we have found that an abnormally high polymer to PCBM weight ratio (1:20) has to be used in stock solution in order to obtain a contact angle in between the two materials. The measured contact angles (95o) on the BHJ films prepared by a weight ratio of 1:20 in both P3HT:PCBM and PTB7:PCBM blend films are approximately in between the values of pure polymer and fullerene. Even at a weight ratio of 1:40, the contact angle of the blend is still 10o higher than that in pure PCBM. Although it is almost impossible to obtain the exact polymer:fullerene


Page 6 of 33

Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


composition ratio with the contact angle, the results unambiguously show that the BHJ films surface which initially covered by polymer has been gradually substituted by PCBM while increasing weight ratio. In addition, the contact angle results are consistent with the phase images of the polymer:PCBM blend acquired by tapping mode atomic force microscopy (AFM) as shown in Figure S1. Clearly separated domains of polymer and PCBM are obtained at high PCBM weight ratio of 1:20.

Figure 1. Optical microscopy of contact angle for polymer:PCBM blend films with weight ratios of (a) P3HT:PCBM in 1:0, 1:20 and 0:1, (b) PTB7:PCBM in 1:0, 1:1 and 1:20, (c) the deionized (DI) water contact angle versus the weight percentage of PCBM in polymer:PCBM blend films.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 shows the UPS spectra of P3HT:PCBM (a, b, and c) and PTB7:PCBM (d, e, and f) blend films with different weight ratios. We have found that only the polymer signal can be detected even up to a weight ratio of 1:10. This agrees with the contact angle results and the previous photoemission reports.36 By further increasing the polymer:fullerene weight ratio, the PCBM signal becomes notable. Taking P3HT:PCBM as an example, the secondary electron cutoff position of pure P3HT film is located at 17.22 eV and the HOMO onset is 0.56 eV below the Fermi level as shown in Figure 2a and c, respectively. Therefore, we have obtained the ionization potential (IP), the energy difference between vacuum level (VL) and HOMO, of 4.56 eV for pure P3HT. This value is in good agreement with that reported by Guan et al..38 Similarly, we have obtained the IP of 6.02 eV for pure PCBM.37,45-47 While increasing the PCBM weight ratio in the blends, we observed a shift of secondary electron cutoff of 0.59 eV towards lower binding energy (BE) (Figure 2a), but the HOMO energy level of P3HT only shifted 0.30 eV towards lower BE (Figure 2c). As a result, the difference is attributed to an interface dipole of 0.29 eV at the P3HT/PCBM interface. On the other hand, for PTB7:PCBM blends, the HOMO energy level of PTB7 is almost the same in different PCBM weight ratios (Figure 2f). Therefore, the shift of 0.14 eV observed in the secondary electron cutoff is only ascribed to the interface dipole at the PTB7/PCBM interface. For both BHJ blends, the HOMO onset of PCBM has no notable shift at different weight ratios (Figure 2b and e).


Page 8 of 33

Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. UPS spectra (He I) of polymer/PCBM interface showing (a, d) the onset of the secondary electron cutoﬀ, (b, e) HOMO edge of PCBM and (c, f) HOMO edge of polymer near the Fermi level with increasing PCBM concentration in blend films. The experimental uncertainty is in ± 0.05 eV.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

In order to confirm the HOMO energy shift in P3HT measured by UPS is not due to charging effect, we compared the UPS results with core energy levels measured by X-ray photoelectron spectroscopy (XPS). Figure 3 shows the XPS core-level spectra of S2p3/2 in P3HT and O1s in PCBM in different weight ratios of blend films. As shown in Figure 3a, a shift of 0.27 eV towards the lower BE was observed in the S2p3/2 peak after mixing P3HT with PCBM at weight ratio of 1:1. By further increasing the weight ratio to 1:80, the same peak shows an additional shift towards even lower BE by 0.10 eV. The total shift of 0.37 eV obtained from the XPS spectra is in good agreement with 0.30 eV deduced from the shift of P3HT HOMO edge measured by UPS. Therefore, the XPS results confirm the charge transfer at the P3HT:PCBM BHJ. Figure 3b shows the O1s spectra of PCBM in different blend films, where the peak at 532.3 eV and 534.0 eV are corresponding to the binding energy of oxygen in carbonyl and methoxy groups respectively.48,49 Different from P3HT, the same BE of O1s indicates a constant Fermi level position in PCBM. Furthermore, neither new component nor peak broadening has been observed in the XPS spectra, which indicates no chemical reaction between P3HT and PCBM. Similarly, Figure 3c and d show the F1s and O1s core level XPS spectra measured in PTB7:PCBM BHJ films with different weight ratios of PCBM, and the results are consistent with the constant HOMO energy level of PTB7 as shown in Figure 2f.


Page 11 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. XPS spectra of (a) S2p3/2, (b) O1s for P3HT:PCBM blend films, (c) F1s and (d) O1s for PTB7:PCBM blend films with different PCBM weight ratios in blend films. The experimental uncertainty is in ± 0.05 eV.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

According to the XPS and UPS results, the energy level diagrams of P3HT:PCBM and PTB7:PCBM can be obtained as shown in Figure 4. The LUMO levels were calculated by considering the measured HOMO levels by UPS and the corresponding charge transporting gaps (2.50 eV for P3HT, 2.25 eV for PTB7 and 2.20 eV for PCBM).39,50 As a result, we obtained the EHOMO-D – ELUMO-A of P3HT:PCBM and PTB7:PCBM equal to 1.03 eV and 1.23 eV, respectively. The origin of the difference of electronic structure between P3HT:PCBM and PTB7:PCBM might be due to the difference in positive energy integer charge transfer state (EICT+) of the two polymers.36,51 The relatively small EICT+ of P3HT has been reported with strong charge transfer to other materials in contact, which possibly facilitates the large energy level shifting.36 Further detailed investigation on the charge transfer states of individual materials would shed light on the underneath mechanism, which is out of the scope of this study. Similarly, the energy level alignments of other polymer:PCBM BHJ blends including MEHPPV:PCBM, PCDTBT:PCBM and PDTSTPD:PCBM were also investigated. The corresponding UPS spectra and energy level diagrams are shown in Figure S2-S4. Consequently, by bringing the PCBM to BHJ film surface with high polymer:fullerene weight ratio in photoemission measurement, it is believed that such direct measurement can provide more reliable correlation between the energy level alignment at the donor/acceptor interfaces (i.e. EHOMO-D – ELUMO-A) and the VOC in photovoltaic devices. Taking P3HT:PCBM as an example, if simply assume VL alignment at the P3HT/PCBM interface, the EHOMO-D – ELUMO-A is only 0.76 eV. However, the measured VOC in photovoltaic devices is around 0.60 - 0.65 V. Therefore, such small value of Eeff in P3HT:PCBM is questionable if we also take the additional VOC loss due to bimolecular recombination into account.27 Moreover, it has also been reported that the VOC of P3HT:PCBM photovoltaic cell can be varied from 0.30 V to 0.95 V by introducing interface


Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dipole layer.52 It suggests that the energy level alignment cannot be simply taken the energy level from individual materials. Otherwise, the Eeff would become too large (1.33 eV) as compared to the VOC (0.6 V) in photovoltaic devices.

Figure 4. Energy diagram of (a) P3HT:PCBM and (b) PTB7:PCBM bulk heterojunctions according to the measured photoemission spectroscopy results shown in Figure 2. All the values are in the unit of eV. The experimental uncertainty is in ± 0.05 eV.

The change in composition of P3HT to PCBM on the surface of blend films with different weight ratios can also be revealed by comparing the atomic ratio of sulfur (S) to carbon (C) from the XPS data as shown in Figure 3. Although oxygen (O) to sulfur atomic ratios can



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

also be used to determine the composition of the blend films, physically absorbed oxygen on the sample surfaces would make the result unreliable. Thus, S/C atomic ratios were used in this study. As shown in Figure 5, the S/C ratio of pure P3HT film is 0.09253 and it decreases to 0.037 with increasing the weight ratio of P3HT:PCBM from 1:1 to 1:80. It is clearly seen that the trend of the atomic ratio obtained by XPS in different weight ratios matches well with the change in contact angles as depicted in Figure 1. It should be noticed that the AFM, contact angle, and XPS composition analysis results are aiming at supporting more fullerene is on the blend surface with the film prepared with high fullerene to polymer ratio. The exact composition of the two materials on surface would require more comprehensive study that is out of the scope of this study. However, such information would be very important to have deeper understanding on the origin of the energy level shifting in BHJ blends.


Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Atomic ratio of S/C for blend films with different weight ratios, the values of S/C taking from the integrated peak area of each element core level in XPS spectra. We sought to compare the measured energy level alignment in polymer:PCBM by PES to the VOC in photovoltaic devices. Recently, as demonstrated by Widmer et al.,26 the extrapolated VOC in various small molecule organic photovoltaic devices at 0 K (maximum attainable VOC, VOC-MAX) is equal to the corresponding Eeff. It agrees with the general empirical expression as shown below:

‫ܸݍ‬௢௖ = ‫ܧ‬௘௙௙ − ݇ܶ ln

ே೎ ேೡ ௡௣

(1)

where the second term on the right-hand side is the temperature dependent photovoltage loss due to bimolecular recombination, q is the elementary charge, k is the Boltzmann constant, Nc and Nv are the density of electronic states in conduction band and valence band, p and n are the hole and electron concentrations, respectively. Figure 6 shows the measured temperature dependent VOC of (a) P3HT:PCBM and (b) PTB7:PCBM photovoltaic devices. The increased VOC at lower temperature is attributed to the reduced bimolecular recombination as described in eq 1, and the VOC-MAX is extrapolated at 0 K which is independent of illumination intensities. Thus, we obtained VOC-MAX of 0.95 ± 0.05 V for P3HT:PCBM and 1.20 ± 0.05 V for PTB7:PCBM. Those values are in excellent agreement with the measured Eeff, 1.03 eV for P3HT:PCBM and 1.23 eV for PTB7:PCBM by the above PES approach. The slightly smaller value of VOC-MAX in P3HT:PCBM is possibly due to the additional loss mechanism, such as small dielectric for poor charge separation as demonstrated by Chen et al..30 The temperature dependent VOC for MEHPPV:PCBM, PCDTBT:PCBM and PDTSTPD:PCBM are shown in Figure S5-S7, the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

corresponding VOC-MAXs also have good agreement with the Eeff determined by PES (Figure S2S4).

Figure 6. Measured temperature dependent VOC of (a) P3HT:PCBM and (b) PTB7:PCBM photovoltaic devices with different illumination intensities.

Concerning the correlation of using high weight ratio in PES measurement and the VOC in photovoltaic devices, we fabricated the devices with high polymer:PCBM ratio and measured the


Page 17 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


photovoltaic characteristic. As shown in Figure S8 in supporting information, the devices with high weight ratio (1:40) have the same VOC as the typical ratio (1:1 and 1:2) used in photovoltaic devices. In fact, such donor:acceptor ratio independent VOC has also been reported in small molecule BHJ devices.26 The same VOC in different weight ratios is possibly because the exciton dissociation process is still mainly occurred at the polymer/fullerene interface. It is different from the case of organic Schottky cells that have been previously reported.54,55 As shown in Figure S9, due to the similar WF for UV-ozone treated indium tin oxide (ITO) substrate (4.90 eV) and PEDOT:PSS (5.10 eV) modified substrate,56 we have found that the measured energy level alignment had negligible substrate effect between using ITO and PEDOT:PSS. It has been demonstrated that VOC of the OPV devices is linearly dependent on the WF difference of the anode and cathode.57 However, as long as the WF difference between the two electrodes is larger than the effective gap, the VOC is independent to the electrode WF difference and solely determined by the Eeff. In our cases, as LiF/Al was used as the cathode (WF = 2.67 eV),58 this gives sufficiently large WF different (>2 eV) compared to the Eeff (

Probing the Energy Level Alignment and the Correlation with Open

Recommend Documents